3dn_xeon_broadwell_patched 2 x Intel Xeon Gold 6140 testing with a Intel S2600WFT (SE5C620.86B.00.01.0013.030920180427 BIOS) and ASPEED ASPEED Family on Debian 9.4 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1805250-FO-3DNXEONBR39&grs .
3dn_xeon_broadwell_patched Processor Motherboard Chipset Memory Disk Graphics Network OS Kernel Compiler File-System Screen Resolution 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 x Intel Xeon E5-2690 v4 @ 2.60GHz (28 Cores / 56 Threads) Intel S2600WTT v.......... (SE5C610.86B.01.01.0024.021320181901 BIOS) Intel Xeon E7 v4/Xeon 8 x 8192 MB DDR4-2134MHz Samsung M393A1G40DB0-CPB 300GB INTEL SSDSC2BB30 + 2000GB INTEL SSDPE2KX020T8 Matrox s MGA G200e [Pilot] Intel 10-Gigabit X540-AT2 Debian 9.4 4.9.0-6-amd64 (x86_64) GCC 6.3.0 20170516 + Clang 3.8.1-24 ext4 1024x768 2 x Intel Xeon Gold 6140 @ 2.30GHz (36 Cores / 72 Threads) Intel S2600WFT (SE5C620.86B.00.01.0013.030920180427 BIOS) Intel Device 2020 8 x 8192 MB DDR4-2133MHz Samsung M393A1G40DB0-CPB 300GB INTEL SSDSC2BB30 + 8002GB INTEL SSDPE2KX080T8 ASPEED ASPEED Family Intel Connection X722 for 10GBASE-T OpenBenchmarking.org Compiler Details - --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic -v Disk Details - none / data=ordered,relatime,rw Processor Details - 2 x Xeon E5-2690 v4: Scaling Governor: intel_pstate performance - 2 x Xeon Gold 6140: Scaling Governor: acpi-cpufreq performance Python Details - Python 2.7.13 + Python 3.5.3 Security Details - KPTI + __user pointer sanitization + Full generic retpoline IBPB IBRS_FW Protection
3dn_xeon_broadwell_patched smallpt: Global Illumination Renderer; 100 Samples iozone: 64Kb - 2GB - Write Performance iozone: 64Kb - 8GB - Write Performance iozone: 64Kb - 4GB - Write Performance cyclictest: clock_nanosleep TIME_ABSTIME, Interval 500 ms, 100000 Loops cyclictest: POSIX Timer, Interval 500 ms, 100000 Loops iozone: 1MB - 512MB - Write Performance llvm-test-suite: Time To Run iozone: 64Kb - 512MB - Write Performance iozone: 4Kb - 512MB - Write Performance fftw: Stock - 1D FFT Size 32 fftw: Stock - 2D FFT Size 32 fio: Seq Write - Linux AIO - No - No - 1MB - /var fio: Rand Write - Linux AIO - No - No - 16KB - /var compilebench: Initial Create fio: Seq Write - Linux AIO - No - Yes - 8MB - /var fio: Rand Write - Linux AIO - No - Yes - 16KB - /var fio: Seq Write - Linux AIO - No - No - 2MB - /var fio: Rand Write - Linux AIO - No - Yes - 512KB - /var glibc-bench: cos glibc-bench: sin blake2: fio: Seq Write - Linux AIO - No - No - 512KB - /var fio: Seq Write - Linux AIO - No - No - 16KB - /var glibc-bench: sqrt fio: Rand Write - Linux AIO - No - Yes - 32KB - /var luajit: Fast Fourier Transform fio: Rand Write - Linux AIO - No - Yes - 64KB - /var fio: Rand Write - Linux AIO - No - Yes - 8MB - /var fio: Rand Write - Linux AIO - No - No - 512KB - /var c-ray: Total Time iozone: 4Kb - 4GB - Read Performance espeak: Text-To-Speech Synthesis primesieve: 1e12 Prime Number Generation fio: Rand Write - Linux AIO - No - Yes - 1MB - /var fio: Seq Write - Linux AIO - No - Yes - 32KB - /var fio: Rand Write - Linux AIO - No - Yes - 128KB - /var himeno: Poisson Pressure Solver fio: Rand Write - Linux AIO - No - Yes - 64MB - /var minion: Quasigroup fio: Seq Write - Linux AIO - No - No - 8MB - /var fio: Rand Write - Linux AIO - No - No - 32KB - /var fio: Seq Write - Linux AIO - No - No - 128KB - /var lzbench: Libdeflate 1 - Compression fio: Rand Write - Linux AIO - No - Yes - 2MB - /var minion: Graceful glibc-bench: pthread_once fio: Rand Write - Linux AIO - No - Yes - 32MB - /var fio: Rand Write - Linux AIO - No - Yes - 4MB - /var fio: Seq Write - Linux AIO - No - No - 16MB - /var glibc-bench: ffsll glibc-bench: ffs fio: Rand Write - Linux AIO - No - No - 64KB - /var fio: Rand Write - Linux AIO - No - No - 128KB - /var redis: SET fio: Seq Write - Linux AIO - No - No - 32MB - /var fio: Seq Write - Linux AIO - No - No - 64MB - /var fio: Seq Write - Linux AIO - No - Yes - 16KB - /var fio: Rand Write - Linux AIO - No - No - 64MB - /var bork: File Encryption Time fio: Rand Write - Linux AIO - No - No - 16MB - /var fio: Rand Write - Linux AIO - No - Yes - 16MB - /var perl-benchmark: Pod2html fio: Rand Write - Linux AIO - No - No - 256KB - /var fio: Seq Write - Linux AIO - No - No - 4MB - /var fio: Rand Write - Linux AIO - No - No - 8MB - /var fio: Rand Write - Linux AIO - No - Yes - 256KB - /var fio: Rand Write - Linux AIO - No - No - 4MB - /var fio: Seq Write - Linux AIO - No - No - 32KB - /var fio: Seq Write - Linux AIO - No - No - 64KB - /var pybench: Total For Average Test Times botan: AES-256 - Encrypt sudokut: Total Time clomp: Static OMP Speedup minion: Solitaire fio: Seq Write - Linux AIO - No - Yes - 128MB - /var fio: Seq Write - Linux AIO - No - No - 256KB - /var fio: Seq Write - Linux AIO - No - Yes - 128KB - /var fio: Seq Write - Linux AIO - No - Yes - 1MB - /var fio: Rand Write - Linux AIO - No - No - 32MB - /var scimark2: Dense LU Matrix Factorization fio: Rand Write - Linux AIO - No - No - 2MB - /var fio: Seq Write - Linux AIO - No - Yes - 16MB - /var iozone: 4Kb - 4GB - Write Performance cython-bench: lzbench: Zstd 1 - Compression redis: SADD fio: Seq Write - Linux AIO - No - Yes - 4MB - /var fio: Seq Write - Linux AIO - No - Yes - 256KB - /var system-libxml2: 112 MB fftw: Stock - 1D FFT Size 64 dcraw: RAW To PPM Image Conversion encode-flac: WAV To FLAC fio: Seq Write - Linux AIO - No - Yes - 512KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 512KB - /var glibc-bench: tanh fio: Rand Write - Linux AIO - No - No - 8KB - /var iozone: 1MB - 8GB - Write Performance fio: Seq Write - Linux AIO - No - Yes - 2MB - /var iozone: 4Kb - 2GB - Write Performance botan: Blowfish - Encrypt fio: Rand Write - Linux AIO - No - No - 128MB - /var fio: Rand Write - Linux AIO - No - Yes - 128MB - /var cachebench: Read / Modify / Write cyclictest: POSIX Timer, Interval 10000 ms, 10000 Loops bullet: Convex Trimesh iozone: 1MB - 2GB - Write Performance openssl: RSA 4096-bit Performance scimark2: Sparse Matrix Multiply john-the-ripper: Blowfish build-llvm: Time To Compile fio: Seq Read - Linux AIO - Yes - No - 512KB - /var fio: Rand Write - Linux AIO - No - Yes - 8KB - /var tjbench: Decompression Throughput fio: Rand Write - Linux AIO - Yes - No - 128MB - /var java-scimark2: Fast Fourier Transform system-libxml2: 1 MB bullet: 1000 Convex redis: LPUSH hint: DOUBLE phpbench: PHP Benchmark Suite bullet: 136 Ragdolls iozone: 1MB - 4GB - Write Performance fio: Seq Write - Linux AIO - No - No - 4KB - /var fio: Seq Write - Linux AIO - No - No - 8KB - /var bullet: Raytests java-scimark2: Sparse Matrix Multiply fio: Rand Write - Linux AIO - No - Yes - 4KB - /var lzbench: Brotli 0 - Compression botan: CAST-256 - Encrypt encode-mp3: WAV To MP3 fio: Seq Write - Linux AIO - No - Yes - 32MB - /var bullet: Prim Trimesh fio: Seq Write - Linux AIO - No - No - 128MB - /var stockfish: Total Time crafty: Elapsed Time botan: Twofish - Encrypt compress-7zip: Compress Speed Test compilebench: Compile build-linux-kernel: Time To Compile fio: Seq Read - Linux AIO - Yes - Yes - 4KB - /var botan: KASUMI - Encrypt fio: Rand Read - Linux AIO - Yes - No - 8KB - /var fio: Seq Write - Linux AIO - No - Yes - 64KB - /var compress-lzma: 256MB File Compression osbench: Create Files build-php: Time To Compile git: Time To Complete Common Git Commands lzbench: XZ 0 - Compression go-benchmark: build cpuminer-opt: sha256t gnupg: 2GB File Encryption fio: Rand Write - Linux AIO - No - No - 4KB - /var bullet: 1000 Stack cachebench: Read hint: FLOAT scimark2: Monte Carlo graphics-magick: Local Adaptive Thresholding bullet: 3000 Fall fio: Seq Write - Linux AIO - No - Yes - 4KB - /var fio: Seq Read - Linux AIO - Yes - No - 4KB - /var java-scimark2: Dense LU Matrix Factorization cachebench: Write mencoder: AVI To LAVC cpuminer-opt: lbry system-decompress-xz: n-queens: Elapsed Time graphics-magick: Sharpen fio: Seq Read - Linux AIO - Yes - Yes - 16MB - /var aio-stress: Rand Write luajit: Sparse Matrix Multiply system-decompress-bzip2: luajit: Dense LU Matrix Factorization postmark: Disk Transaction Performance rbenchmark: scimark2: Composite java-scimark2: Composite fio: Seq Write - Linux AIO - No - Yes - 8KB - /var compress-gzip: Linux Source Tree Archiving To .tar.gz cpuminer-opt: skein scikit-learn: fio: Seq Read - Linux AIO - Yes - Yes - 64MB - /var fio: Seq Write - Linux AIO - No - Yes - 64MB - /var nginx: Static Web Page Serving fs-mark: 1000 Files, 1MB Size, No Sync/FSync graphics-magick: Blur build-gcc: Time To Compile cpuminer-opt: m7m graphics-magick: Resizing john-the-ripper: Traditional DES fio: Seq Read - Linux AIO - Yes - Yes - 1MB - /var unpack-linux: linux-4.15.tar.xz asmfish: 1024 Hash Memory, 26 Depth fio: Seq Read - Linux AIO - Yes - No - 2MB - /var luajit: Monte Carlo osbench: Create Threads tscp: AI Chess Performance aobench: 2048 x 2048 - Total Time fio: Seq Write - Linux AIO - Yes - No - 32KB - /var dbench: 1 encode-wavpack: WAV To WavPack graphics-magick: HWB Color Space compilebench: Read Compiled Tree fio: Seq Read - Linux AIO - Yes - No - 64MB - /var fio: Seq Write - Linux AIO - Yes - No - 512KB - /var luajit: Composite fio: Rand Write - Linux AIO - Yes - Yes - 64KB - /var fio: Rand Write - Linux AIO - Yes - Yes - 4KB - /var fio: Seq Read - Linux AIO - No - No - 4KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 16KB - /var iozone: 4Kb - 512MB - Read Performance fio: Rand Read - Linux AIO - Yes - Yes - 8KB - /var fio: Seq Read - Linux AIO - Yes - No - 128MB - /var fio: Seq Read - Linux AIO - Yes - Yes - 4MB - /var fs-mark: 1000 Files, 1MB Size fio: Seq Read - Linux AIO - Yes - No - 4MB - /var fio: Rand Read - Linux AIO - Yes - No - 512KB - /var fio: Seq Read - Linux AIO - Yes - No - 8MB - /var fio: Seq Read - Linux AIO - No - Yes - 4KB - /var go-benchmark: json ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping build-imagemagick: Time To Compile sample-program: fs-mark: 4000 Files, 32 Sub Dirs, 1MB Size fio: Rand Read - Linux AIO - No - Yes - 4KB - /var fio: Rand Read - Linux AIO - Yes - No - 128KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 32MB - /var fio: Seq Read - Linux AIO - Yes - Yes - 128MB - /var fio: Rand Read - Linux AIO - Yes - No - 64KB - /var scimark2: Fast Fourier Transform iozone: 1MB - 2GB - Read Performance fhourstones: Complex Connect-4 Solving iozone: 64Kb - 2GB - Read Performance fio: Seq Read - Linux AIO - No - No - 16KB - /var hmmer: Pfam Database Search t-test1: 1 scimark2: Jacobi Successive Over-Relaxation fio: Rand Read - Linux AIO - Yes - Yes - 1MB - /var perl-benchmark: Interpreter fio: Rand Read - Linux AIO - Yes - Yes - 16MB - /var fio: Rand Read - Linux AIO - No - No - 4KB - /var iozone: 64Kb - 4GB - Read Performance fio: Rand Read - Linux AIO - Yes - No - 2MB - /var t-test1: 2 dbench: 6 osbench: Memory Allocations fio: Rand Read - Linux AIO - Yes - Yes - 4MB - /var dbench: 12 fio: Rand Read - Linux AIO - Yes - No - 8MB - /var java-scimark2: Jacobi Successive Over-Relaxation fio: Rand Write - Linux AIO - Yes - Yes - 4MB - /var fio: Rand Write - Linux AIO - Yes - No - 32MB - /var fio: Rand Read - Linux AIO - Yes - Yes - 128MB - /var x264: H.264 Video Encoding java-scimark2: Monte Carlo fio: Rand Read - Linux AIO - Yes - No - 64MB - /var blogbench: Write apache: Static Web Page Serving fio: Rand Read - Linux AIO - Yes - No - 4MB - /var fio: Rand Read - Linux AIO - Yes - No - 32MB - /var java-gradle-perf: Reactor fio: Seq Read - Linux AIO - No - No - 2MB - /var ramspeed: Triad - Integer fio: Rand Read - Linux AIO - Yes - No - 16KB - /var xsbench: fio: Rand Read - Linux AIO - Yes - Yes - 64MB - /var blogbench: Read fio: Rand Read - Linux AIO - Yes - No - 128MB - /var fio: Rand Read - Linux AIO - Yes - No - 16MB - /var fio: Seq Read - Linux AIO - No - Yes - 128MB - /var fio: Seq Write - Linux AIO - Yes - No - 128MB - /var fio: Seq Read - Linux AIO - No - No - 128KB - /var fio: Seq Read - Linux AIO - No - Yes - 256KB - /var fio: Seq Read - Linux AIO - No - No - 128MB - /var multichase: 1GB Array, 256 Byte Stride ramspeed: Scale - Integer fio: Rand Read - Linux AIO - No - Yes - 128KB - /var fio: Seq Read - Linux AIO - No - No - 4MB - /var fio: Rand Read - Linux AIO - No - No - 128KB - /var go-benchmark: garbage hackbench: 32 - Process dbench: 128 stream: Add fio: Rand Read - Linux AIO - No - Yes - 128MB - /var stream: Scale hackbench: 8 - Thread fio: Seq Read - Linux AIO - No - No - 16MB - /var dbench: 48 stream: Triad fio: Seq Read - Linux AIO - No - No - 256KB - /var fio: Rand Read - Linux AIO - No - Yes - 256KB - /var luajit: Jacobi Successive Over-Relaxation fio: Rand Read - Linux AIO - No - No - 32MB - /var fio: Seq Read - Linux AIO - No - Yes - 512KB - /var fio: Rand Read - Linux AIO - No - No - 64MB - /var fio: Rand Read - Linux AIO - No - No - 16MB - /var fio: Seq Read - Linux AIO - No - Yes - 64MB - /var fio: Rand Read - Linux AIO - No - Yes - 1MB - /var fio: Rand Read - Linux AIO - No - No - 256KB - /var fio: Seq Read - Linux AIO - No - Yes - 64KB - /var fio: Rand Read - Linux AIO - No - Yes - 2MB - /var fio: Seq Read - Linux AIO - No - No - 32MB - /var fio: Rand Read - Linux AIO - No - No - 16KB - /var fio: Rand Read - Linux AIO - No - No - 128MB - /var multichase: 1GB Array, 256 Byte Stride, 2 Threads fio: Rand Read - Linux AIO - No - Yes - 16KB - /var fio: Seq Read - Linux AIO - No - Yes - 32KB - /var fio: Rand Read - Linux AIO - No - Yes - 4MB - /var fio: Rand Read - Linux AIO - No - Yes - 64MB - /var fio: Rand Read - Linux AIO - No - No - 8KB - /var fio: Rand Read - Linux AIO - No - Yes - 8MB - /var ramspeed: Average - Integer tensorflow: Cifar10 ramspeed: Copy - Integer fio: Seq Read - Linux AIO - No - Yes - 32MB - /var fio: Seq Read - Linux AIO - No - No - 32KB - /var fio: Rand Read - Linux AIO - No - Yes - 512KB - /var fio: Seq Read - Linux AIO - No - No - 8MB - /var fio: Seq Read - Linux AIO - No - No - 64KB - /var dbench: 256 fs-mark: 5000 Files, 1MB Size, 4 Threads fio: Seq Read - Linux AIO - No - Yes - 4MB - /var fio: Seq Read - Linux AIO - No - Yes - 8KB - /var fio: Seq Read - Linux AIO - No - Yes - 8MB - /var fio: Rand Read - Linux AIO - No - Yes - 32MB - /var fio: Rand Read - Linux AIO - No - Yes - 8KB - /var povray: Trace Time fio: Seq Read - Linux AIO - No - No - 512KB - /var fio: Rand Read - Linux AIO - No - Yes - 64KB - /var fio: Rand Read - Linux AIO - No - No - 64KB - /var go-benchmark: http fio: Rand Read - Linux AIO - No - No - 512KB - /var fio: Seq Read - Linux AIO - No - Yes - 16KB - /var fio: Rand Read - Linux AIO - No - No - 8MB - /var fio: Rand Read - Linux AIO - No - No - 32KB - /var sqlite: Timed SQLite Insertions ramspeed: Add - Floating Point fio: Rand Read - Linux AIO - No - Yes - 32KB - /var stream: Copy fio: Seq Read - Linux AIO - No - Yes - 2MB - /var multichase: 4MB Array, 64 Byte Stride ramspeed: Triad - Floating Point ramspeed: Copy - Floating Point multichase: 256MB Array, 256 Byte Stride fio: Seq Read - Linux AIO - No - No - 64MB - /var fio: Seq Read - Linux AIO - No - Yes - 16MB - /var ramspeed: Scale - Floating Point multichase: 1GB Array, 256 Byte Stride, 4 Threads fio: Seq Read - Linux AIO - No - No - 8KB - /var fio: Rand Read - Linux AIO - No - No - 1MB - /var fio: Rand Read - Linux AIO - No - Yes - 16MB - /var osbench: Launch Programs fio: Rand Read - Linux AIO - No - No - 2MB - /var y-cruncher: Calculating 500M Pi Digits ramspeed: Average - Floating Point ramspeed: Add - Integer fio: Rand Read - Linux AIO - No - No - 4MB - /var cyclictest: clock_nanosleep TIME_ABSTIME, Interval 10000 ms, 10000 Loops hackbench: 32 - Thread ffmpeg: H.264 HD To NTSC DV compress-pbzip2: 256MB File Compression vpxenc: vpxenc hpcg: fio: Seq Read - Linux AIO - No - Yes - 1MB - /var osbench: Create Processes sunflow: Global Illumination + Image Synthesis redis: GET redis: LPOP hackbench: 16 - Process hackbench: 8 - Process hackbench: 4 - Process hackbench: 2 - Process hackbench: 16 - Thread hackbench: 1 - Process hackbench: 4 - Thread hackbench: 2 - Thread hackbench: 1 - Thread ebizzy: cpuminer-opt: cryptonight cpuminer-opt: myr-gr cpuminer-opt: deep iozone: 64Kb - 512MB - Read Performance iozone: 1MB - 512MB - Read Performance iozone: 64Kb - 8GB - Read Performance iozone: 4Kb - 8GB - Write Performance iozone: 4Kb - 8GB - Read Performance iozone: 4Kb - 2GB - Read Performance iozone: 1MB - 8GB - Read Performance iozone: 1MB - 4GB - Read Performance fio: Seq Write - Linux AIO - Yes - Yes - 512KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 256KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 128MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 128KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 64MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 64KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 32MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 32KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 16MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 16KB - /var fio: Seq Write - Linux AIO - Yes - No - 256KB - /var fio: Seq Write - Linux AIO - Yes - No - 128KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 256KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 128KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 8MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 8KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 4MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 4KB - /var fio: Seq Write - Linux AIO - Yes - Yes - 2MB - /var fio: Seq Write - Linux AIO - Yes - Yes - 1MB - /var fio: Seq Write - Linux AIO - Yes - No - 64MB - /var fio: Seq Write - Linux AIO - Yes - No - 64KB - /var fio: Seq Write - Linux AIO - Yes - No - 32MB - /var fio: Seq Write - Linux AIO - Yes - No - 16MB - /var fio: Seq Write - Linux AIO - Yes - No - 16KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 64KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 32KB - /var fio: Seq Read - Linux AIO - Yes - No - 256KB - /var fio: Seq Read - Linux AIO - Yes - No - 128KB - /var fio: Seq Read - Linux AIO - No - Yes - 128KB - /var fio: Seq Write - Linux AIO - Yes - No - 8MB - /var fio: Seq Write - Linux AIO - Yes - No - 8KB - /var fio: Seq Write - Linux AIO - Yes - No - 4MB - /var fio: Seq Write - Linux AIO - Yes - No - 4KB - /var fio: Seq Write - Linux AIO - Yes - No - 2MB - /var fio: Seq Write - Linux AIO - Yes - No - 1MB - /var fio: Seq Read - Linux AIO - Yes - Yes - 8MB - /var fio: Seq Read - Linux AIO - Yes - Yes - 8KB - /var fio: Seq Read - Linux AIO - Yes - Yes - 2MB - /var fio: Seq Read - Linux AIO - Yes - No - 64KB - /var fio: Seq Read - Linux AIO - Yes - No - 32MB - /var fio: Seq Read - Linux AIO - Yes - No - 32KB - /var fio: Seq Read - Linux AIO - Yes - No - 16MB - /var fio: Seq Read - Linux AIO - Yes - No - 16KB - /var fio: Seq Read - Linux AIO - Yes - No - 8KB - /var fio: Seq Read - Linux AIO - Yes - No - 1MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 512KB - /var fio: Rand Write - Linux AIO - Yes - Yes - 256KB - /var fio: Rand Write - Linux AIO - Yes - Yes - 128MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 128KB - /var fio: Seq Read - Linux AIO - No - No - 1MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 64MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 32MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 32KB - /var fio: Rand Write - Linux AIO - Yes - Yes - 16MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 16KB - /var fio: Rand Write - Linux AIO - Yes - No - 512KB - /var fio: Rand Write - Linux AIO - Yes - No - 256KB - /var fio: Rand Write - Linux AIO - Yes - No - 128KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 512KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 256KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 128KB - /var fio: Rand Write - Linux AIO - Yes - Yes - 8MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 8KB - /var fio: Rand Write - Linux AIO - Yes - Yes - 2MB - /var fio: Rand Write - Linux AIO - Yes - Yes - 1MB - /var fio: Rand Write - Linux AIO - Yes - No - 64MB - /var fio: Rand Write - Linux AIO - Yes - No - 64KB - /var fio: Rand Write - Linux AIO - Yes - No - 32KB - /var fio: Rand Write - Linux AIO - Yes - No - 16MB - /var fio: Rand Write - Linux AIO - Yes - No - 16KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 64KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 32MB - /var fio: Rand Read - Linux AIO - Yes - Yes - 32KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 16KB - /var fio: Rand Read - Linux AIO - Yes - No - 256KB - /var fio: Rand Write - Linux AIO - Yes - No - 8MB - /var fio: Rand Write - Linux AIO - Yes - No - 8KB - /var fio: Rand Write - Linux AIO - Yes - No - 4MB - /var fio: Rand Write - Linux AIO - Yes - No - 4KB - /var fio: Rand Write - Linux AIO - Yes - No - 2MB - /var fio: Rand Write - Linux AIO - Yes - No - 1MB - /var fio: Rand Read - Linux AIO - Yes - Yes - 8MB - /var fio: Rand Read - Linux AIO - Yes - Yes - 4KB - /var fio: Rand Read - Linux AIO - Yes - Yes - 2MB - /var fio: Rand Read - Linux AIO - Yes - No - 32KB - /var fio: Rand Write - Linux AIO - No - No - 1MB - /var fio: Rand Read - Linux AIO - Yes - No - 4KB - /var fio: Rand Read - Linux AIO - Yes - No - 1MB - /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 15 565.88 627.21 567.41 2 2 616.85 475.70 650.70 594.72 5355.20 5544.43 1797 1753 300.86 1843 1826 1841 1844 149070 149069 6.31 1842 1859 17.18 1875 311.48 1881 1901 1894 4.65 3140.96 50.43 16.44 1913 1896 1871 1576.12 2046 192.60 1941 1929 1881 126 1933 89.10 7.03 1997 1943 1966 6.02 6.02 1934 1909 1029397.88 2005 2065 1942 2078 12.86 1989 1993 0.19414052 1939 1975 1982 1951 1979 1968 1967 2287 2231.02 17.94 1.33 118.01 2243 1953 1942 1955 2061 586.26 2007 2031 713.60 43.71 237 1180993.44 2001 1965 104960 5224.77 68.37 14.44 2015 3397 49.86 1663 897.44 1991 730.76 169.31 2334 2338 28417.19 3 1.80 809.28 5320.23 450.65 31756 183.79 3440 1618 107.90 2195 918.84 707 7.15 1043663.50 551045311.25 321808 4.53 830.84 942 1747 4.00 1449.62 921 259 85.69 47.00 2118 1.49 2405 4995 5278790 221.90 102477 1149.90 40.81 1905 57.24 1464 1975 374.88 22.64 66.37 8.67 23 18084527804 38.05 19.57 935 7.73 2199.95 253586598.16 89.63 65 6.56 939 1954 3867.54 18374.10 28.05 53.95 5.55 7.34 147 3284 1796.28 858.35 15.10 2327.48 3906 0.6377 442.58 1633.78 1803 53.87 63.10 54.80 3425 2014 23188.82 1301.57 123 735.33 553.98 157 118189333 3367 8.03 61125168 3361 348.48 22.08 901142 67.35 2348 740.72 10.35 166 811.01 3558 2343 1029.56 2354 1215 979 3948 3976.10 1346 3764 3484 742.63 3497 3445 3516 1002 3321327 614.05 32.10 20.28 755.17 1013 5480 3591 3894 5235 222.23 5492.67 10165.67 5820.38 1862 8.02 25.75 864.12 3310 0.00149524 3302 995 6074.30 3342 8.45 3457.21 113.34 3361 5373.54 3414 1143.85 2404 2395 3749 122.35 789.03 3615 33732 24725.96 3490 3558 24.95 2582 21066.88 2790 4712538 3673 1503221 3873 3543 2974 2685 2833 2514 3020 71.84 19586.19 2801 2510 2836 1102350 32.60 7997.28 79281.28 3087 71203.52 11.22 2813 8171.60 80592.48 2593 2537 1302.00 2705 2511 2787 2684 2711 2581 2578 2488 2582 2646 2139 3156 76.71 2134 2416 2581 2762 1760 2594 20288.79 34.47 19682.27 2620 2431 2523 2491 2463 8087.02 1516.80 2508 1809 2520 2656 1778 27.62 2565 2420 2424 8031 2572 2024 2684 2289 2.07 21799.18 2297 81634.76 2493 9.50 21730.75 19182.32 71.57 2685 2577 19611.31 82.12 1594 3066 2615 50.15 2921 16.29 20567.55 21596.76 2736 2 33.76 9.08 2.44 1.02 0.95 2486 42.43 0.97 1377549.17 1518694.56 17.94 10.47 5.86 3.77 19.10 2.98 6.07 4.13 3.27 623000 630.06 6238.62 16.02 5546.56 4989.07 5525.86 650.55 3798.10 3234.92 4700.61 5618.88 2160 2433 2467 2361 2042 1833 2272 2355 2143 2038 2421 2285 4307 5077 2251 1458 2056 1255 2066 2017 2199 2049 2101 1937 1975 5310 4525 4490 5211 2820 1932 1793 2103 1348 2291 2117 3329 2754 3280 4507 3399 4250 3419 3853 2850 3192 2004 2082 2234 2400 2827 1984 2155 2056 2356 2011 1724 2108 2139 3246 4748 5280 2218 1492 1972 2148 2486 1895 2110 2330 1797 4716 3324 4007 2683 4287 2086 1273 2199 1094 2070 2068 3317 57.00 3336 3865 1927 159 3251 141 1194.14 1320.24 1194.15 1 1 1176.60 250.70 1171.65 1068.42 9492.85 9749.30 3110 3007 515.62 3128 3096 3114 3111 88796.57 88811.97 3.77 3081 3109 10.32 3113 515.28 3104 3135 3111 2.84 5142.61 30.94 10.10 3112 3082 3039 2551.96 3311 119.13 3135 3112 3033 203 3114 55.34 4.37 3211 3122 3158 3.75 3.75 3103 3055 1646931.25 3205 3295 3098 3311 8.09 3159 3160 0.12257338 3070 3122 3133 3080 3123 3104 3098 1454 3502.25 11.45 0.85 75.42 3508 3052 3032 3048 3211 912.87 3114 3151 1103.45 28.27 365 1818448.15 3081 3022 68397 8017.40 44.59 9.45 3079 5172 32.77 2530 1363.77 3023 1102.37 255.30 3511 3512 42686.23 2 1.20 1212.72 7953.37 671.92 47328 123.39 5122 2407 160.47 3258 1359.96 478 4.84 1539273.58 811930300.26 473325 3.08 1220.40 1381 2560 2.73 2122.19 1348 379 125.28 32.15 3096 1.02 3509 3428 7668552 321.93 148413 1664.62 28.25 2748 82.54 1017 2842 260.64 15.75 46.18 6.04 33 12606362535 54.51 13.69 1336 5.43 3127.90 360166182.52 126.94 92 4.64 1326 2758 5453.19 25887.13 19.92 75.67 3.98 5.27 204 4556 2490.95 1189.02 10.92 3217.62 5396 0.4632 608.68 2238.26 2470 39.34 86.31 40.10 4673 2739 31509.79 1760.60 166 545.00 746.66 211 158775333 4523 6.01 81661129 4478 463.18 16.67 1193201 50.89 3107 979.89 7.84 219 1069.68 4689 3086 1355.45 3083 1588 1278 5153 5175.24 1035 4894 4502 951.63 4475 4398 4476 1275 2615724 778.86 25.40 16.06 945.30 1268 6851 4488 4863 6519 275.40 6788.91 12512.47 7160.35 2288 6.53 21.02 1056.26 4042 0.00122757 4007 1207 7330.93 4027 7.03 4152.44 95.45 3982 6357.75 4022 1345.40 2827 2044 4380 141.58 910.53 4167 38841 28462.09 3987 4064 21.85 2948 24040.99 3183 5364296 4177 1330002 4367 3986 3334 2396 3142 2784 3333 79.28 21571.85 3084 2738 3089 1199814 30.02 7367.04 85973.52 3331 76762.86 10.41 3031 7593.44 86517.06 2780 2719 1392.17 2531 2673 2623 2528 2556 2434 2729 2351 2440 2799 2026 3330 80.85 2025 2293 2451 2623 1672 2465 21327.26 32.82 18773.03 2499 2548 2642 2607 2576 7736.44 1585.47 2402 1736 2419 2551 1713 26.63 2660 2340 2345 8297 2651 1969 2614 2234 2.12 22325.64 2243 79721.98 2439 9.70 21301.65 18838.58 72.86 2732 2537 19834.23 83.02 1611 3093 2593 49.74 2910 16.23 20641.24 21564.59 2734 2 0.95 2486 40.02 0.87 2063823.71 2265203.02 19.11 10.06 7.49 5.30 18.14 3.29 6.92 5.42 3.17 943906 796.30 11.93 19.01 6005.59 6830.55 7000.87 1201.49 5165.63 5078.78 6607.57 5329.01 2114 2876 2971 3029 2609 2718 2954 2484 2310 2831 2853 2914 6227 7210 2903 2146 2334 1677 2932 2605 2640 3104 2964 2378 2337 6393 6092 6903 7109 3138 2474 2319 2326 1693 2229 2545 4518 4104 4465 6783 4572 6093 4534 5125 4086 4475 2542 2828 2357 2604 3149 2402 2696 2885 2428 2327 2688 2813 2894 4436 6306 6735 2851 1879 2370 2055 2719 3058 2639 2563 2391 4055 4057 4615 3272 6517 2354 1849 2712 1386 2484 2466 4017 31.73 4029 4920 3102 31.83 4043 OpenBenchmarking.org
Smallpt Global Illumination Renderer; 100 Samples OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 100 Samples 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 2.33, N = 3 15 141 1. (CXX) g++ options: -fopenmp
IOzone Record Size: 64Kb - File Size: 2GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 2GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 7.74, N = 3 SE +/- 3.22, N = 3 565.88 1194.14 1. (CC) gcc options: -O3
IOzone Record Size: 64Kb - File Size: 8GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 8GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 0.56, N = 3 SE +/- 6.01, N = 3 627.21 1320.24 1. (CC) gcc options: -O3
IOzone Record Size: 64Kb - File Size: 4GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 4GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 7.33, N = 3 SE +/- 2.53, N = 3 567.41 1194.15 1. (CC) gcc options: -O3
Cyclictest Test Case: clock_nanosleep TIME_ABSTIME, Interval 500 ms, 100000 Loops OpenBenchmarking.org ms Average, Fewer Is Better Cyclictest 0.84 Test Case: clock_nanosleep TIME_ABSTIME, Interval 500 ms, 100000 Loops 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.45 0.9 1.35 1.8 2.25 2 1 1. (CC) gcc options: -O2 -lrt -lpthread -lrttest
Cyclictest Test Case: POSIX Timer, Interval 500 ms, 100000 Loops OpenBenchmarking.org ms Average, Fewer Is Better Cyclictest 0.84 Test Case: POSIX Timer, Interval 500 ms, 100000 Loops 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.45 0.9 1.35 1.8 2.25 2 1 1. (CC) gcc options: -O2 -lrt -lpthread -lrttest
IOzone Record Size: 1MB - File Size: 512MB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 11.17, N = 6 SE +/- 3.37, N = 3 616.85 1176.60 1. (CC) gcc options: -O3
LLVM Test Suite Time To Run OpenBenchmarking.org Seconds, Fewer Is Better LLVM Test Suite 6.0.0 Time To Run 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 100 200 300 400 500 SE +/- 0.15, N = 3 SE +/- 0.42, N = 3 475.70 250.70 1. (CXX) g++ options: -O3 -lstdc++
IOzone Record Size: 64Kb - File Size: 512MB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 7.56, N = 3 SE +/- 14.19, N = 3 650.70 1171.65 1. (CC) gcc options: -O3
IOzone Record Size: 4Kb - File Size: 512MB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 6.61, N = 3 SE +/- 15.79, N = 3 594.72 1068.42 1. (CC) gcc options: -O3
FFTW Build: Stock - Size: 1D FFT Size 32 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 32 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 20.92, N = 3 SE +/- 178.58, N = 6 5355.20 9492.85 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
FFTW Build: Stock - Size: 2D FFT Size 32 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 32 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 5.74, N = 3 SE +/- 34.85, N = 3 5544.43 9749.30 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 28.75, N = 3 SE +/- 0.88, N = 3 1797 3110 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 26.81, N = 6 SE +/- 12.90, N = 3 1753 3007 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Compile Bench Test: Initial Create OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 110 220 330 440 550 SE +/- 5.33, N = 3 SE +/- 2.41, N = 3 300.86 515.62
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 1855.92, N = 6 SE +/- 881.92, N = 3 112333 192333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 16.64, N = 3 SE +/- 4.36, N = 3 1843 3128 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 25.49, N = 6 SE +/- 1.20, N = 3 1826 3096 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 1673.32, N = 6 117000 198000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 21.39, N = 3 1841 3114 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 25.27, N = 3 SE +/- 2.85, N = 3 1844 3111 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
glibc bench Benchmark: cos OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: cos 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 37.00, N = 3 SE +/- 30.58, N = 3 149070.00 88796.57
glibc bench Benchmark: sin OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: sin 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 26.54, N = 3 SE +/- 10.00, N = 3 149069.00 88811.97
BLAKE2 OpenBenchmarking.org Cycles Per Byte, Fewer Is Better BLAKE2 20170307 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 6.31 3.77 1. (CC) gcc options: -O3 -march=native -lcrypto -lz
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 31.20, N = 4 SE +/- 4.91, N = 3 1842 3081 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 31.62, N = 6 1859 3109 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 2033.06, N = 6 119000 199000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
glibc bench Benchmark: sqrt OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: sqrt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 17.18 10.32
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 20.95, N = 3 SE +/- 0.58, N = 3 1875 3113 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 676.59, N = 3 60033 99600 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
LuaJIT Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.0.5 Test: Fast Fourier Transform 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 110 220 330 440 550 SE +/- 0.27, N = 3 SE +/- 3.32, N = 3 311.48 515.28 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 29.78, N = 3 1881 3104 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11K 22K 33K 44K 55K SE +/- 472.58, N = 3 SE +/- 33.33, N = 3 30100 49667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 39.13, N = 6 1901 3135 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 37.34, N = 6 SE +/- 0.58, N = 3 1894 3111 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
C-Ray Total Time OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.0463 2.0926 3.1389 4.1852 5.2315 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 4.65 2.84 1. (CC) gcc options: -lm -lpthread -O3
IOzone Record Size: 4Kb - File Size: 4GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 4GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 12.47, N = 3 SE +/- 16.67, N = 3 3140.96 5142.61 1. (CC) gcc options: -O3
eSpeak Speech Engine Text-To-Speech Synthesis OpenBenchmarking.org Seconds, Fewer Is Better eSpeak Speech Engine 1.48.04 Text-To-Speech Synthesis 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11 22 33 44 55 SE +/- 0.08, N = 3 SE +/- 0.19, N = 3 50.43 30.94 1. (CXX) g++ options: -lstdc++ -lespeak -O2 -fpic -fvisibility=hidden -pedantic -fno-exceptions
Primesieve 1e12 Prime Number Generation OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 6.2 1e12 Prime Number Generation 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 16.44 10.10 1. (CXX) g++ options: -O3 -rdynamic -lpthread
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 43.81, N = 6 SE +/- 1.76, N = 3 1913 3112 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 1129.23, N = 6 SE +/- 120.19, N = 3 60650 98633 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 35.33, N = 6 SE +/- 3.48, N = 3 1896 3082 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 252.65, N = 6 14950 24300 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 31.60, N = 6 SE +/- 3.28, N = 3 1871 3039 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 344.38, N = 5 SE +/- 88.19, N = 3 15060 24433 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Himeno Benchmark Poisson Pressure Solver OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 4.10, N = 3 SE +/- 2.24, N = 3 1576.12 2551.96 1. (CC) gcc options: -O3 -mavx2
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 39.08, N = 6 2046 3311 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Minion Benchmark: Quasigroup OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Quasigroup 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 SE +/- 0.56, N = 3 SE +/- 0.24, N = 3 192.60 119.13 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 34.83, N = 3 1941 3135 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 405.52, N = 3 61733 99600 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 11.86, N = 3 1929 3112 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 37.43, N = 3 SE +/- 2.73, N = 3 1881 3033 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 272.85, N = 3 SE +/- 33.33, N = 3 15033 24233 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
lzbench Test: Libdeflate 1 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Libdeflate 1 - Process: Compression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 126 203 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 29.48, N = 5 1933 3114 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Minion Benchmark: Graceful OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Graceful 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 89.10 55.34 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
glibc bench Benchmark: pthread_once OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: pthread_once 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 7.03 4.37
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 17.01, N = 3 SE +/- 0.33, N = 3 1997 3211 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 22.82, N = 3 1943 3122 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 39.62, N = 6 1966 3158 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
glibc bench Benchmark: ffsll OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: ffsll 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 6.02 3.75
glibc bench Benchmark: ffs OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: ffs 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 6.02 3.75
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11K 22K 33K 44K 55K SE +/- 417.67, N = 3 SE +/- 33.33, N = 3 30933 49633 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 26.39, N = 3 SE +/- 1.00, N = 3 1934 3103 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 28.46, N = 5 SE +/- 12.86, N = 3 1909 3055 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Redis Test: SET OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400K 800K 1200K 1600K 2000K SE +/- 18402.25, N = 6 SE +/- 35144.79, N = 6 1029397.88 1646931.25 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 37.55, N = 6 SE +/- 1.33, N = 3 2005 3205 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 333.33, N = 3 SE +/- 333.33, N = 3 124333 198667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 35.32, N = 6 SE +/- 10.87, N = 3 2065 3295 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 8.50, N = 3 SE +/- 5.03, N = 3 1942 3098 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 37.59, N = 6 2078 3311 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Bork File Encrypter File Encryption Time OpenBenchmarking.org Seconds, Fewer Is Better Bork File Encrypter 1.4 File Encryption Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 3 6 9 12 15 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 12.86 8.09
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 32.32, N = 6 SE +/- 0.67, N = 3 1989 3159 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 352.77, N = 3 SE +/- 233.33, N = 3 62667 99367 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 12.57, N = 3 1993 3160 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Perl Benchmarks Test: Pod2html OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Pod2html 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.0437 0.0874 0.1311 0.1748 0.2185 SE +/- 0.00017341, N = 3 SE +/- 0.00012724, N = 3 0.19414052 0.12257338
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 11.02, N = 3 SE +/- 2.31, N = 3 1939 3070 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 1.53, N = 3 1975 3122 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 11.24, N = 3 SE +/- 0.88, N = 3 1982 3133 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 2.65, N = 3 SE +/- 4.93, N = 3 1951 3080 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 10.02, N = 3 1979 3123 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 3.76, N = 3 SE +/- 7.69, N = 3 1968 3104 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Botan Test: AES-256 - Encrypt - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: AES-256 - Encrypt - Decrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 800 1600 2400 3200 4000 SE +/- 4.32, N = 3 SE +/- 2.63, N = 3 2219.58 3498.68 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 2.33, N = 3 SE +/- 4.37, N = 3 1967 3098 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11K 22K 33K 44K 55K SE +/- 33.33, N = 3 SE +/- 66.67, N = 3 31467 49533 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
PyBench Total For Average Test Times OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 3.28, N = 3 SE +/- 3.76, N = 3 2287 1454
Botan Test: AES-256 - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: AES-256 - Encrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 800 1600 2400 3200 4000 SE +/- 0.80, N = 3 SE +/- 0.20, N = 3 2231.02 3502.25 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Sudokut Total Time OpenBenchmarking.org Seconds, Fewer Is Better Sudokut 0.4 Total Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.00, N = 3 17.94 11.45
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 208.17, N = 3 SE +/- 66.67, N = 3 15500 24267 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
CLOMP Static OMP Speedup OpenBenchmarking.org Speedup, More Is Better CLOMP 3.3 Static OMP Speedup 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.2993 0.5986 0.8979 1.1972 1.4965 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 1.33 0.85 1. (CC) gcc options: --openmp -O3 -lm
Minion Benchmark: Solitaire OpenBenchmarking.org Seconds, Fewer Is Better Minion 1.8 Benchmark: Solitaire 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 0.30, N = 3 SE +/- 0.49, N = 3 118.01 75.42 1. (CXX) g++ options: -std=gnu++11 -O3 -fomit-frame-pointer -rdynamic
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 800 1600 2400 3200 4000 SE +/- 39.55, N = 3 SE +/- 3.00, N = 3 2243 3508 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 35.70, N = 3 SE +/- 6.67, N = 3 1953 3052 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 26.52, N = 3 SE +/- 7.36, N = 3 1942 3032 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 39.75, N = 6 SE +/- 21.67, N = 3 1955 3048 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 20.26, N = 3 2061 3211 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 0.28, N = 3 SE +/- 0.63, N = 3 586.26 912.87 1. (CC) gcc options: -lm
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 1.76, N = 3 SE +/- 1.33, N = 3 2007 3114 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 5.69, N = 3 SE +/- 0.88, N = 3 2031 3151 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
IOzone Record Size: 4Kb - File Size: 4GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 4GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 11.28, N = 6 SE +/- 0.29, N = 3 713.60 1103.45 1. (CC) gcc options: -O3
Cython benchmark OpenBenchmarking.org Seconds, Fewer Is Better Cython benchmark 0.27 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 10 20 30 40 50 SE +/- 0.33, N = 3 SE +/- 0.27, N = 3 43.71 28.27
lzbench Test: Zstd 1 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Zstd 1 - Process: Compression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 80 160 240 320 400 SE +/- 1.00, N = 3 SE +/- 0.33, N = 3 237 365 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Redis Test: SADD OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SADD 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400K 800K 1200K 1600K 2000K SE +/- 28725.74, N = 6 SE +/- 43613.03, N = 6 1180993.44 1818448.15 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 6.96, N = 3 SE +/- 34.84, N = 3 2001 3081 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 20.28, N = 3 SE +/- 2.52, N = 3 1965 3022 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
System Libxml2 Parsing Filesize: 112 MB OpenBenchmarking.org ms, Fewer Is Better System Libxml2 Parsing Filesize: 112 MB 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 1888.21, N = 3 SE +/- 603.82, N = 3 104960 68397
FFTW Build: Stock - Size: 1D FFT Size 64 OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 1D FFT Size 64 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 4.63, N = 3 SE +/- 103.35, N = 3 5224.77 8017.40 1. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
dcraw RAW To PPM Image Conversion OpenBenchmarking.org Seconds, Fewer Is Better dcraw RAW To PPM Image Conversion 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 15 30 45 60 75 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 68.37 44.59 1. (CC) gcc options: -lm
FLAC Audio Encoding WAV To FLAC OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4 8 12 16 20 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 14.44 9.45 1. (CXX) g++ options: -O2 -fvisibility=hidden -lm
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 3.71, N = 3 SE +/- 11.14, N = 3 2015 3079 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 2185.81, N = 3 SE +/- 2645.75, N = 3 212667 324000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 64.23, N = 6 SE +/- 10.02, N = 3 3397 5172 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
glibc bench Benchmark: tanh OpenBenchmarking.org nanoseconds, Fewer Is Better glibc bench 1.0 Benchmark: tanh 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11 22 33 44 55 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 49.86 32.77
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 17.16, N = 3 SE +/- 21.23, N = 3 1663 2530 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
IOzone Record Size: 1MB - File Size: 8GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 8GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 15.20, N = 3 SE +/- 3.20, N = 3 897.44 1363.77 1. (CC) gcc options: -O3
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 32.65, N = 4 SE +/- 15.84, N = 3 1991 3023 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Botan Test: Blowfish - Encrypt - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: Blowfish - Encrypt - Decrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 60 120 180 240 300 SE +/- 0.04, N = 3 SE +/- 0.12, N = 3 167.79 254.67 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
IOzone Record Size: 4Kb - File Size: 2GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 2GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 11.34, N = 3 SE +/- 0.47, N = 3 730.76 1102.37 1. (CC) gcc options: -O3
Botan Test: Blowfish - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: Blowfish - Encrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 60 120 180 240 300 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 169.31 255.30 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 800 1600 2400 3200 4000 SE +/- 41.82, N = 3 SE +/- 0.58, N = 3 2334 3511 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 800 1600 2400 3200 4000 SE +/- 14.84, N = 3 2338 3512 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
CacheBench Test: Read / Modify / Write OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read / Modify / Write 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 9K 18K 27K 36K 45K SE +/- 10.11, N = 3 SE +/- 2.73, N = 3 28417.19 42686.23 1. (CC) gcc options: -lrt
Cyclictest Test Case: POSIX Timer, Interval 10000 ms, 10000 Loops OpenBenchmarking.org ms Average, Fewer Is Better Cyclictest 0.84 Test Case: POSIX Timer, Interval 10000 ms, 10000 Loops 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.675 1.35 2.025 2.7 3.375 3 2 1. (CC) gcc options: -O2 -lrt -lpthread -lrttest
Bullet Physics Engine Test: Convex Trimesh OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Convex Trimesh 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.405 0.81 1.215 1.62 2.025 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 1.80 1.20 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
IOzone Record Size: 1MB - File Size: 2GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 2GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 13.23, N = 3 SE +/- 5.00, N = 3 809.28 1212.72 1. (CC) gcc options: -O3
OpenSSL RSA 4096-bit Performance OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.0f RSA 4096-bit Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 1.07, N = 3 SE +/- 7.46, N = 3 5320.23 7953.37 1. (CC) gcc options: -O3 -pthread -m64 -lssl -lcrypto -ldl
SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 140 280 420 560 700 SE +/- 0.03, N = 3 SE +/- 0.19, N = 3 450.65 671.92 1. (CC) gcc options: -lm
John The Ripper Test: Blowfish OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.8.0-jumbo-1 Test: Blowfish 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 10K 20K 30K 40K 50K SE +/- 5.33, N = 3 SE +/- 103.11, N = 3 31756 47328 1. (CC) gcc options: -lssl -lcrypto -fopenmp -pthread -lm -lz -ldl -lcrypt
Timed LLVM Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 4.0.1 Time To Compile 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 SE +/- 2.68, N = 3 SE +/- 0.34, N = 3 183.79 123.39
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 6.94, N = 3 SE +/- 24.37, N = 3 3440 5122 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 26.51, N = 3 SE +/- 40.50, N = 3 1618 2407 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
libjpeg-turbo tjbench Test: Decompression Throughput OpenBenchmarking.org Megapixels/sec, More Is Better libjpeg-turbo tjbench 1.5.3 Test: Decompression Throughput 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 SE +/- 0.39, N = 3 SE +/- 0.37, N = 3 107.90 160.47 1. (CC) gcc options: -O3 -lm
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 3382.96, N = 3 SE +/- 5131.60, N = 3 207333 308000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 5.69, N = 3 SE +/- 4.41, N = 3 2195 3258 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
lzbench Test: Zstd 1 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Zstd 1 - Process: Decompression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 0.58, N = 3 653 969 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Java SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Fast Fourier Transform 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 3.33, N = 4 SE +/- 5.13, N = 4 918.84 1359.96
System Libxml2 Parsing Filesize: 1 MB OpenBenchmarking.org ms, Fewer Is Better System Libxml2 Parsing Filesize: 1 MB 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 150 300 450 600 750 SE +/- 3.67, N = 3 SE +/- 0.88, N = 3 707 478
Bullet Physics Engine Test: 1000 Convex OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 7.15 4.84 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Redis Test: LPUSH OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPUSH 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300K 600K 900K 1200K 1500K SE +/- 9796.56, N = 3 SE +/- 4172.99, N = 3 1043663.50 1539273.58 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Hierarchical INTegration Test: DOUBLE OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: DOUBLE 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200M 400M 600M 800M 1000M SE +/- 281862.65, N = 3 SE +/- 810117.24, N = 3 551045311.25 811930300.26 1. (CC) gcc options: -O3 -march=native -lm
Botan Test: Twofish - Encrypt - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: Twofish - Encrypt - Decrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70 140 210 280 350 SE +/- 0.06, N = 3 SE +/- 0.25, N = 3 219.90 323.53 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
PHPBench PHP Benchmark Suite OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 100K 200K 300K 400K 500K SE +/- 1354.78, N = 3 SE +/- 416.21, N = 3 321808 473325
Bullet Physics Engine Test: 136 Ragdolls OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.0193 2.0386 3.0579 4.0772 5.0965 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 4.53 3.08 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
IOzone Record Size: 1MB - File Size: 4GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 4GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 5.62, N = 3 SE +/- 2.41, N = 3 830.84 1220.40 1. (CC) gcc options: -O3
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 80K 160K 240K 320K 400K SE +/- 3511.88, N = 3 SE +/- 666.67, N = 3 241000 353667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 14.11, N = 3 SE +/- 2.85, N = 3 942 1381 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 36.62, N = 6 SE +/- 22.81, N = 3 1747 2560 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Bullet Physics Engine Test: Raytests OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Raytests 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.9 1.8 2.7 3.6 4.5 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 4.00 2.73 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 4550.95, N = 6 SE +/- 2848.00, N = 3 223667 327667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Java SciMark Computational Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Sparse Matrix Multiply 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 13.92, N = 4 SE +/- 0.43, N = 4 1449.62 2122.19
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 3062.32, N = 6 SE +/- 1000.00, N = 3 235667 345000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 11.90, N = 6 SE +/- 3.67, N = 3 921 1348 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
lzbench Test: Brotli 0 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Brotli 0 - Process: Compression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 80 160 240 320 400 259 379 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Botan Test: CAST-256 - Encrypt - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: CAST-256 - Encrypt - Decrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 85.65 125.29 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Botan Test: CAST-256 - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: CAST-256 - Encrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 85.69 125.28 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
LAME MP3 Encoding WAV To MP3 OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11 22 33 44 55 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 47.00 32.15 1. (CC) gcc options: -lm
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 5.36, N = 3 SE +/- 37.49, N = 3 2118 3096 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Bullet Physics Engine Test: Prim Trimesh OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Prim Trimesh 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.3353 0.6706 1.0059 1.3412 1.6765 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.49 1.02 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 800 1600 2400 3200 4000 SE +/- 4.98, N = 3 SE +/- 0.67, N = 3 2405 3509 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Stockfish Total Time OpenBenchmarking.org ms, Fewer Is Better Stockfish 2014-11-26 Total Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 6.43, N = 3 SE +/- 2.65, N = 3 4995 3428 1. (CXX) g++ options: -lpthread -fno-exceptions -fno-rtti -ansi -pedantic -O3 -msse -msse3 -mpopcnt -flto
Crafty Elapsed Time OpenBenchmarking.org Nodes Per Second, More Is Better Crafty 25.2 Elapsed Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.6M 3.2M 4.8M 6.4M 8M SE +/- 2766.70, N = 3 SE +/- 2436.96, N = 3 5278790 7668552
Botan Test: Twofish - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: Twofish - Encrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70 140 210 280 350 SE +/- 0.05, N = 3 SE +/- 0.22, N = 3 221.90 321.93 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
lzbench Test: Libdeflate 1 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Libdeflate 1 - Process: Decompression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 0.33, N = 3 686 994 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
7-Zip Compression Compress Speed Test OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 683.20, N = 3 SE +/- 852.32, N = 3 102477 148413 1. (CXX) g++ options: -pipe -lpthread
Compile Bench Test: Compile OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 7.03, N = 3 SE +/- 2.23, N = 3 1149.90 1664.62
Timed Linux Kernel Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 4.13 Time To Compile 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 9 18 27 36 45 SE +/- 0.61, N = 5 SE +/- 0.38, N = 6 40.81 28.25
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 37.84, N = 6 SE +/- 3.79, N = 3 1905 2748 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 4121.08, N = 6 187500 130000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Botan Test: KASUMI - Encrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: KASUMI - Encrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 57.24 82.54 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 150K 300K 450K 600K 750K SE +/- 9631.26, N = 6 SE +/- 881.92, N = 3 487833 703333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
lzbench Test: Brotli 0 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: Brotli 0 - Process: Decompression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 110 220 330 440 550 SE +/- 0.58, N = 3 342 493 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 31.54, N = 6 SE +/- 1.86, N = 3 1464 1017 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 10K 20K 30K 40K 50K SE +/- 437.16, N = 3 SE +/- 176.38, N = 3 31567 45433 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 27.49, N = 3 SE +/- 11.32, N = 3 1975 2842 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
LZMA Compression 256MB File Compression OpenBenchmarking.org Seconds, Fewer Is Better LZMA Compression 256MB File Compression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 80 160 240 320 400 SE +/- 0.20, N = 3 SE +/- 0.36, N = 3 374.88 260.64 1. (CXX) g++ options: -O2
OSBench Test: Create Files OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 22.64 15.75 1. (CC) gcc options: -lm
Timed PHP Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.1.9 Time To Compile 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 15 30 45 60 75 SE +/- 0.27, N = 3 SE +/- 0.13, N = 3 66.37 46.18 -lz 1. (CC) gcc options: -O2 -pedantic -ldl -lm
Botan Test: KASUMI - Encrypt - Decrypt OpenBenchmarking.org MiB/s, More Is Better Botan 2.4.0 Test: KASUMI - Encrypt - Decrypt 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 55.39 79.57 1. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt
Git Time To Complete Common Git Commands OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 8.67 6.04 1. git version 2.11.0
lzbench Test: XZ 0 - Process: Compression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: XZ 0 - Process: Compression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 8 16 24 32 40 23 33 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
Go Benchmarks Test: build OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: build 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4000M 8000M 12000M 16000M 20000M SE +/- 124732337.55, N = 3 SE +/- 190660145.77, N = 3 18084527804 12606362535
Cpuminer-Opt Algorithm: sha256t OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: sha256t 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12 24 36 48 60 SE +/- 0.57, N = 6 SE +/- 0.63, N = 3 38.05 54.51 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
GnuPG 2GB File Encryption OpenBenchmarking.org Seconds, Fewer Is Better GnuPG 1.4.22 2GB File Encryption 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 19.57 13.69 1. (CC) gcc options: -O2 -MT -MD -MP -MF
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 666.67, N = 3 SE +/- 1000.00, N = 3 239333 342000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 3.18, N = 3 SE +/- 3.93, N = 3 935 1336 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Bullet Physics Engine Test: 1000 Stack OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 7.73 5.43 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
CacheBench Test: Read OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Read 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 0.00, N = 3 SE +/- 0.15, N = 3 2199.95 3127.90 1. (CC) gcc options: -lrt
Hierarchical INTegration Test: FLOAT OpenBenchmarking.org QUIPs, More Is Better Hierarchical INTegration 1.0 Test: FLOAT 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 80M 160M 240M 320M 400M SE +/- 127357.80, N = 3 SE +/- 87821.79, N = 3 253586598.16 360166182.52 1. (CC) gcc options: -O3 -march=native -lm
SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 0.19, N = 3 SE +/- 0.03, N = 3 89.63 126.94 1. (CC) gcc options: -lm
GraphicsMagick Operation: Local Adaptive Thresholding OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Local Adaptive Thresholding 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 65 92 -ljbig -ltiff -ljpeg -llzma 1. (CC) gcc options: -fopenmp -O2 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lgomp -lpthread
Bullet Physics Engine Test: 3000 Fall OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.12, N = 3 SE +/- 0.03, N = 3 6.56 4.64 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 2027.59, N = 3 SE +/- 4522.17, N = 6 240333 339500 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 150K 300K 450K 600K 750K SE +/- 6992.46, N = 6 SE +/- 2000.00, N = 3 499833 706000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 7.22, N = 3 SE +/- 17.69, N = 6 939 1326 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 27.44, N = 6 SE +/- 8.02, N = 3 1954 2758 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Java SciMark Computational Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Dense LU Matrix Factorization 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1200 2400 3600 4800 6000 SE +/- 37.74, N = 4 SE +/- 36.41, N = 4 3867.54 5453.19
CacheBench Test: Write OpenBenchmarking.org MB/s, More Is Better CacheBench Test: Write 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6K 12K 18K 24K 30K SE +/- 55.46, N = 3 SE +/- 91.99, N = 3 18374.10 25887.13 1. (CC) gcc options: -lrt
Mencoder AVI To LAVC OpenBenchmarking.org Seconds, Fewer Is Better Mencoder 1.3.0 AVI To LAVC 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 7 14 21 28 35 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 28.05 19.92 -lasound -lsndio 1. (CC) gcc options: -ffast-math -fpie -pie -lrt -lpng -lz -ljpeg -ldl -lpthread -lbz2 -rdynamic -lm
Cpuminer-Opt Algorithm: lbry OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: lbry 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.52, N = 3 SE +/- 0.89, N = 3 53.95 75.67 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
System XZ Decompression OpenBenchmarking.org Seconds, Fewer Is Better System XZ Decompression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.2488 2.4976 3.7464 4.9952 6.244 SE +/- 0.09, N = 4 SE +/- 0.00, N = 3 5.55 3.98
lzbench Test: XZ 0 - Process: Decompression OpenBenchmarking.org MB/s, More Is Better lzbench 2017-08-08 Test: XZ 0 - Process: Decompression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 66 92 1. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
N-Queens Elapsed Time OpenBenchmarking.org Seconds, Fewer Is Better N-Queens 1.0 Elapsed Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 7.34 5.27 1. (CC) gcc options: -static -fopenmp -O3 -march=native
GraphicsMagick Operation: Sharpen OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Sharpen 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 147 204 -ljbig -ltiff -ljpeg -llzma 1. (CC) gcc options: -fopenmp -O2 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lgomp -lpthread
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 78.87, N = 6 SE +/- 9.70, N = 3 3284 4556 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
AIO-Stress Test: Random Write OpenBenchmarking.org MB/s, More Is Better AIO-Stress 0.21 Test: Random Write 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 24.17, N = 3 SE +/- 16.89, N = 3 1796.28 2490.95 1. (CC) gcc options: -pthread -laio
LuaJIT Test: Sparse Matrix Multiply OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.0.5 Test: Sparse Matrix Multiply 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 0.19, N = 3 SE +/- 1.08, N = 3 858.35 1189.02 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
System BZIP2 Decompression OpenBenchmarking.org Seconds, Fewer Is Better System BZIP2 Decompression 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 15.10 10.92
LuaJIT Test: Dense LU Matrix Factorization OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.0.5 Test: Dense LU Matrix Factorization 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 4.36, N = 3 SE +/- 16.70, N = 3 2327.48 3217.62 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
PostMark Disk Transaction Performance OpenBenchmarking.org TPS, More Is Better PostMark 1.51 Disk Transaction Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1200 2400 3600 4800 6000 SE +/- 38.33, N = 3 3906 5396 1. (CC) gcc options: -O3
R Benchmark OpenBenchmarking.org Seconds, Fewer Is Better R Benchmark 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.1435 0.287 0.4305 0.574 0.7175 SE +/- 0.0030, N = 3 SE +/- 0.0019, N = 3 0.6377 0.4632 1. R scripting front-end version 3.3.3 (2017-03-06)
SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 130 260 390 520 650 SE +/- 0.05, N = 3 SE +/- 0.60, N = 3 442.58 608.68 1. (CC) gcc options: -lm
Java SciMark Computational Test: Composite OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Composite 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 9.54, N = 4 SE +/- 7.88, N = 4 1633.78 2238.26
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 3282.95, N = 3 SE +/- 3605.55, N = 3 230667 316000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 26.62, N = 3 SE +/- 27.55, N = 3 1803 2470 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Gzip Compression Linux Source Tree Archiving To .tar.gz OpenBenchmarking.org Seconds, Fewer Is Better Gzip Compression Linux Source Tree Archiving To .tar.gz 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12 24 36 48 60 SE +/- 0.02, N = 3 SE +/- 0.19, N = 3 53.87 39.34
Cpuminer-Opt Algorithm: skein OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: skein 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.73, N = 3 SE +/- 0.05, N = 3 63.10 86.31 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Scikit-Learn OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.17.1 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12 24 36 48 60 SE +/- 0.58, N = 3 SE +/- 0.17, N = 3 54.80 40.10
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 82.95, N = 6 SE +/- 23.40, N = 3 3425 4673 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 22.98, N = 3 SE +/- 48.44, N = 3 2014 2739 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
NGINX Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 7K 14K 21K 28K 35K SE +/- 143.98, N = 3 SE +/- 45.68, N = 3 23188.82 31509.79 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
FS-Mark Test: 1000 Files, 1MB Size, No Sync/FSync OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size, No Sync/FSync 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 15.30, N = 3 SE +/- 5.95, N = 3 1301.57 1760.60 1. (CC) gcc options: -static
GraphicsMagick Operation: Blur OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Blur 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 123 166 -ljbig -ltiff -ljpeg -llzma 1. (CC) gcc options: -fopenmp -O2 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lgomp -lpthread
Timed GCC Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed GCC Compilation 7.2 Time To Compile 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 160 320 480 640 800 SE +/- 0.32, N = 3 SE +/- 0.27, N = 3 735.33 545.00
Cpuminer-Opt Algorithm: m7m OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: m7m 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 160 320 480 640 800 SE +/- 0.22, N = 3 SE +/- 5.56, N = 3 553.98 746.66 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
GraphicsMagick Operation: Resizing OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: Resizing 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50 100 150 200 250 SE +/- 0.33, N = 3 157 211 -ljbig -ltiff -ljpeg -llzma 1. (CC) gcc options: -fopenmp -O2 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lgomp -lpthread
John The Ripper Test: Traditional DES OpenBenchmarking.org Real C/S, More Is Better John The Ripper 1.8.0-jumbo-1 Test: Traditional DES 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30M 60M 90M 120M 150M SE +/- 46634.51, N = 3 SE +/- 952698.68, N = 3 118189333 158775333 1. (CC) gcc options: -lssl -lcrypto -fopenmp -pthread -lm -lz -ldl -lcrypt
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 66.32, N = 6 SE +/- 16.64, N = 3 3367 4523 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Unpacking The Linux Kernel linux-4.15.tar.xz OpenBenchmarking.org Seconds, Fewer Is Better Unpacking The Linux Kernel linux-4.15.tar.xz 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.01, N = 4 SE +/- 0.04, N = 4 8.03 6.01
asmFish 1024 Hash Memory, 26 Depth OpenBenchmarking.org Nodes/second, More Is Better asmFish 2017-09-19 1024 Hash Memory, 26 Depth 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20M 40M 60M 80M 100M SE +/- 638460.27, N = 3 SE +/- 767584.86, N = 3 61125168 81661129
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 73.50, N = 6 SE +/- 26.03, N = 3 3361 4478 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
LuaJIT Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.0.5 Test: Monte Carlo 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 100 200 300 400 500 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 348.48 463.18 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OSBench Test: Create Threads OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.44, N = 6 SE +/- 0.24, N = 5 22.08 16.67 1. (CC) gcc options: -lm
TSCP AI Chess Performance OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300K 600K 900K 1200K 1500K SE +/- 294.40, N = 5 SE +/- 1550.50, N = 5 901142 1193201 1. (CC) gcc options: -O3 -march=native
AOBench Size: 2048 x 2048 - Total Time OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 15 30 45 60 75 SE +/- 0.09, N = 3 SE +/- 0.02, N = 3 67.35 50.89 1. (CC) gcc options: -lm -O3
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 5.46, N = 3 SE +/- 7.42, N = 3 2348 3107 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 176.38, N = 3 SE +/- 251.66, N = 3 75133 99400 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Dbench Client Count: 1 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 1 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 7.90, N = 3 SE +/- 7.91, N = 3 740.72 979.89 1. (CC) gcc options: -lpopt -O2
WavPack Audio Encoding WAV To WavPack OpenBenchmarking.org Seconds, Fewer Is Better WavPack Audio Encoding 5.1 WAV To WavPack 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 3 6 9 12 15 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 10.35 7.84 1. (CC) gcc options: -O2 -lm
GraphicsMagick Operation: HWB Color Space OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.28 Operation: HWB Color Space 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50 100 150 200 250 SE +/- 1.20, N = 3 166 219 -ljbig -ltiff -ljpeg -llzma 1. (CC) gcc options: -fopenmp -O2 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lgomp -lpthread
Compile Bench Test: Read Compiled Tree OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Read Compiled Tree 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 7.42, N = 3 SE +/- 9.61, N = 3 811.01 1069.68
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 83.47, N = 6 SE +/- 12.90, N = 3 3558 4689 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 7.64, N = 3 SE +/- 5.20, N = 3 2343 3086 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
LuaJIT Test: Composite OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.0.5 Test: Composite 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 0.98, N = 3 SE +/- 4.21, N = 3 1029.56 1355.45 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 4.70, N = 3 SE +/- 10.15, N = 3 2354 3083 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11K 22K 33K 44K 55K SE +/- 66.67, N = 3 SE +/- 152.75, N = 3 37667 49300 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 90K 180K 270K 360K 450K SE +/- 1527.53, N = 3 SE +/- 1201.85, N = 3 311000 406667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 2185.81, N = 3 SE +/- 666.67, N = 3 250333 327333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 5.61, N = 3 SE +/- 4.37, N = 3 1215 1588 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 666.67, N = 3 SE +/- 2886.75, N = 3 252667 330000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 8.02, N = 3 SE +/- 1.86, N = 3 979 1278 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 11.29, N = 3 SE +/- 41.90, N = 3 3948 5153 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
IOzone Record Size: 4Kb - File Size: 512MB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 512MB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 0.38, N = 3 SE +/- 30.83, N = 3 3976.10 5175.24 1. (CC) gcc options: -O3
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 3850.69, N = 6 SE +/- 1855.92, N = 3 172167 132333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 30.50, N = 6 SE +/- 14.86, N = 3 1346 1035 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 80.58, N = 6 SE +/- 3.18, N = 3 3764 4894 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 3.53, N = 3 SE +/- 13.62, N = 3 3484 4502 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
FS-Mark Test: 1000 Files, 1MB Size OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 10.92, N = 3 SE +/- 3.61, N = 3 742.63 951.63 1. (CC) gcc options: -static
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 7.22, N = 3 SE +/- 11.46, N = 3 3497 4475 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 13.65, N = 3 SE +/- 24.27, N = 3 3445 4398 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 11.14, N = 3 SE +/- 16.15, N = 3 3516 4476 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 7.42, N = 3 SE +/- 1.86, N = 3 1002 1275 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 1855.92, N = 3 SE +/- 333.33, N = 3 256667 326333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Go Benchmarks Test: json OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: json 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700K 1400K 2100K 2800K 3500K SE +/- 8949.35, N = 3 SE +/- 11620.22, N = 3 3321327 2615724
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3a Phong Rendering With Soft-Shadow Mapping 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 0.91, N = 3 SE +/- 1.72, N = 3 614.05 778.86 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lstdc++
Timed ImageMagick Compilation Time To Compile OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 7 14 21 28 35 SE +/- 0.49, N = 3 SE +/- 0.44, N = 3 32.10 25.40
Sample Pi Program OpenBenchmarking.org Seconds, Fewer Is Better Sample Pi Program 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 20.28 16.06
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12K 24K 36K 48K 60K SE +/- 266.67, N = 3 43500 54767 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 4200.53, N = 6 SE +/- 333.33, N = 3 259333 324667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
FS-Mark Test: 4000 Files, 32 Sub Dirs, 1MB Size OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 1.95, N = 3 SE +/- 0.49, N = 3 755.17 945.30 1. (CC) gcc options: -static
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 16.06, N = 6 1013 1268 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 22.92, N = 3 SE +/- 33.69, N = 3 5480 6851 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 5.81, N = 3 SE +/- 72.30, N = 3 3591 4488 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 9.45, N = 3 SE +/- 21.01, N = 3 3894 4863 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 504.42, N = 3 SE +/- 1201.85, N = 3 83733 104333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 32.27, N = 3 SE +/- 89.71, N = 3 5235 6519 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
SciMark Computational Test: Fast Fourier Transform OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 60 120 180 240 300 SE +/- 0.28, N = 3 SE +/- 2.45, N = 3 222.23 275.40 1. (CC) gcc options: -lm
IOzone Record Size: 1MB - File Size: 2GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 2GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 4.38, N = 3 SE +/- 62.53, N = 3 5492.67 6788.91 1. (CC) gcc options: -O3
Fhourstones Complex Connect-4 Solving OpenBenchmarking.org Kpos / sec, More Is Better Fhourstones 3.1 Complex Connect-4 Solving 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 3K 6K 9K 12K 15K SE +/- 25.45, N = 3 SE +/- 18.90, N = 3 10165.67 12512.47 1. (CC) gcc options: -O3
IOzone Record Size: 64Kb - File Size: 2GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 2GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 19.77, N = 3 SE +/- 34.76, N = 3 5820.38 7160.35 1. (CC) gcc options: -O3
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 14.22, N = 3 SE +/- 22.58, N = 3 1862 2288 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Timed HMMer Search Pfam Database Search OpenBenchmarking.org Seconds, Fewer Is Better Timed HMMer Search 2.3.2 Pfam Database Search 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 8.02 6.53 1. (CC) gcc options: -O2 -pthread -lhmmer -lsquid -lm
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 881.92, N = 3 SE +/- 1452.97, N = 3 119333 146333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
t-test1 Threads: 1 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 1 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6 12 18 24 30 SE +/- 0.05, N = 3 SE +/- 0.20, N = 3 25.75 21.02 1. (CC) gcc options: -pthread
SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 864.12 1056.26 1. (CC) gcc options: -lm
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 78.86, N = 6 SE +/- 10.71, N = 3 3310 4042 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Perl Benchmarks Test: Interpreter OpenBenchmarking.org Seconds, Fewer Is Better Perl Benchmarks Test: Interpreter 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.0003 0.0006 0.0009 0.0012 0.0015 SE +/- 0.00002405, N = 3 SE +/- 0.00000173, N = 3 0.00149524 0.00122757
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 79.41, N = 6 SE +/- 22.91, N = 3 3302 4007 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 17.69, N = 6 SE +/- 1.86, N = 3 995 1207 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 4585.61, N = 6 SE +/- 577.35, N = 3 254833 309000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
IOzone Record Size: 64Kb - File Size: 4GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 4GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1600 3200 4800 6400 8000 SE +/- 39.85, N = 3 SE +/- 49.79, N = 3 6074.30 7330.93 1. (CC) gcc options: -O3
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 75.68, N = 6 SE +/- 17.35, N = 3 3342 4027 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
t-test1 Threads: 2 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 2 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 8.45 7.03 1. (CC) gcc options: -pthread
Dbench Client Count: 6 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 6 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 17.01, N = 3 SE +/- 2.48, N = 3 3457.21 4152.44 1. (CC) gcc options: -lpopt -O2
OSBench Test: Memory Allocations OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 0.92, N = 3 SE +/- 0.40, N = 3 113.34 95.45 1. (CC) gcc options: -lm
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 78.78, N = 6 SE +/- 7.51, N = 3 3361 3982 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Dbench Client Count: 12 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 12 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 27.52, N = 3 SE +/- 5.72, N = 3 5373.54 6357.75 1. (CC) gcc options: -lpopt -O2
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 68.95, N = 6 SE +/- 5.21, N = 3 3414 4022 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Java SciMark Computational Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 1.10, N = 4 SE +/- 0.14, N = 4 1143.85 1345.40
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 6.36, N = 3 SE +/- 13.43, N = 3 2404 2827 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 9.94, N = 3 SE +/- 5.36, N = 3 2395 2044 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 82.81, N = 6 SE +/- 12.50, N = 3 3749 4380 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
x264 H.264 Video Encoding OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-02-05 H.264 Video Encoding 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30 60 90 120 150 SE +/- 1.41, N = 3 SE +/- 2.34, N = 4 122.35 141.58 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
Java SciMark Computational Test: Monte Carlo OpenBenchmarking.org Mflops, More Is Better Java SciMark 2.0 Computational Test: Monte Carlo 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 8.90, N = 4 SE +/- 0.25, N = 4 789.03 910.53
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 62.66, N = 6 SE +/- 12.67, N = 3 3615 4167 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
BlogBench Test: Write OpenBenchmarking.org Final Score, More Is Better BlogBench 1.0 Test: Write 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 8K 16K 24K 32K 40K SE +/- 270.44, N = 3 SE +/- 471.13, N = 3 33732 38841 1. (CC) gcc options: -O2 -pthread
Apache Benchmark Static Web Page Serving OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6K 12K 18K 24K 30K SE +/- 186.34, N = 3 SE +/- 314.09, N = 3 24725.96 28462.09 1. (CC) gcc options: -shared -fPIC -O2 -pthread
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 6.56, N = 3 SE +/- 6.23, N = 3 3490 3987 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 3.84, N = 3 SE +/- 5.78, N = 3 3558 4064 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Java Gradle Build Gradle Build: Reactor OpenBenchmarking.org Seconds, Fewer Is Better Java Gradle Build 1.0 Gradle Build: Reactor 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6 12 18 24 30 SE +/- 0.16, N = 4 SE +/- 0.18, N = 4 24.95 21.85
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 14.15, N = 3 SE +/- 38.14, N = 3 2582 2948 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
RAMspeed SMP Type: Triad - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Triad - Benchmark: Integer 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K 21066.88 24040.99 1. (CC) gcc options: -O3 -march=native
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 4318.56, N = 6 SE +/- 1666.67, N = 3 178500 203667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 66.54, N = 6 SE +/- 26.64, N = 3 2790 3183 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Xsbench OpenBenchmarking.org Lookups/s, More Is Better Xsbench 2017-07-06 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.1M 2.2M 3.3M 4.4M 5.5M SE +/- 12567.27, N = 3 SE +/- 16599.98, N = 3 4712538 5364296 1. (CC) gcc options: -std=gnu99 -fopenmp -O3 -lm
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 3.51, N = 3 SE +/- 8.62, N = 3 3673 4177 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
BlogBench Test: Read OpenBenchmarking.org Final Score, More Is Better BlogBench 1.0 Test: Read 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300K 600K 900K 1200K 1500K SE +/- 28598.60, N = 3 SE +/- 24342.41, N = 3 1503221 1330002 1. (CC) gcc options: -O2 -pthread
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 1.20, N = 3 SE +/- 15.59, N = 3 3873 4367 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 8.09, N = 3 SE +/- 14.77, N = 3 3543 3986 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 16.46, N = 3 SE +/- 34.64, N = 3 2974 3334 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.40, N = 3 SE +/- 50.65, N = 6 2685 2396 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 64.25, N = 6 SE +/- 34.12, N = 3 2833 3142 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 513.59, N = 6 SE +/- 272.85, N = 3 22667 25133 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 9.91, N = 3 SE +/- 24.83, N = 3 2514 2784 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 11.98, N = 3 SE +/- 36.73, N = 3 3020 3333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.69, N = 3 SE +/- 0.42, N = 3 71.84 79.28 1. (CC) gcc options: -O2 -static -pthread -lrt
RAMspeed SMP Type: Scale - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Scale - Benchmark: Integer 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K 19586.19 21571.85 1. (CC) gcc options: -O3 -march=native
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 46.54, N = 3 SE +/- 49.43, N = 3 2801 3084 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.86, N = 3 SE +/- 23.97, N = 3 2510 2738 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 43.88, N = 3 SE +/- 48.81, N = 3 2836 3089 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Go Benchmarks Test: garbage OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: garbage 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300K 600K 900K 1200K 1500K SE +/- 11832.84, N = 3 SE +/- 8893.10, N = 3 1102350 1199814
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 346.41, N = 3 SE +/- 404.15, N = 3 22700 24700 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Hackbench Count: 32 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 32 - Type: Process 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 8 16 24 32 40 SE +/- 0.17, N = 3 SE +/- 0.09, N = 3 32.60 30.02 1. (CC) gcc options: -lpthread
Dbench Client Count: 128 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 128 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 4.18, N = 3 SE +/- 8.52, N = 3 7997.28 7367.04 1. (CC) gcc options: -lpopt -O2
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 375.65, N = 3 SE +/- 688.80, N = 3 22433 24333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Stream Type: Add OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 311.66, N = 5 SE +/- 16.65, N = 5 79281.28 85973.52 1. (CC) gcc options: -O3 -march=native -fopenmp
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 40.68, N = 3 SE +/- 35.83, N = 3 3087 3331 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Stream Type: Scale OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 16K 32K 48K 64K 80K SE +/- 222.00, N = 5 SE +/- 8.71, N = 5 71203.52 76762.86 1. (CC) gcc options: -O3 -march=native -fopenmp
Hackbench Count: 8 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Thread 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 3 6 9 12 15 SE +/- 0.18, N = 4 SE +/- 0.16, N = 3 11.22 10.41 1. (CC) gcc options: -lpthread
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 52.05, N = 3 SE +/- 39.89, N = 3 2813 3031 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Dbench Client Count: 48 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 48 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 0.97, N = 3 SE +/- 15.67, N = 3 8171.60 7593.44 1. (CC) gcc options: -lpopt -O2
Stream Type: Triad OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 253.37, N = 5 SE +/- 7.92, N = 5 80592.48 86517.06 1. (CC) gcc options: -O3 -march=native -fopenmp
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 33.33, N = 3 SE +/- 88.19, N = 3 10133 10867 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 5.04, N = 3 SE +/- 23.97, N = 3 2593 2780 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 4.67, N = 3 SE +/- 24.54, N = 3 2537 2719 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
LuaJIT Test: Jacobi Successive Over-Relaxation OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.0.5 Test: Jacobi Successive Over-Relaxation 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 0.09, N = 3 SE +/- 0.16, N = 3 1302.00 1392.17 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 3.53, N = 3 SE +/- 2.96, N = 3 2705 2531 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 3.48, N = 3 SE +/- 19.92, N = 3 2511 2673 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.40, N = 3 SE +/- 1.45, N = 3 2787 2623 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.67, N = 3 SE +/- 9.54, N = 3 2684 2528 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.73, N = 3 SE +/- 2.52, N = 3 2711 2556 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 0.58, N = 3 SE +/- 1.45, N = 3 2581 2434 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 6.49, N = 3 SE +/- 26.27, N = 3 2578 2729 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 1.00, N = 3 SE +/- 3.51, N = 3 2488 2351 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 2582 2440 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 5.24, N = 3 SE +/- 19.97, N = 3 2646 2799 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 9K 18K 27K 36K 45K SE +/- 33.33, N = 3 39800 37633 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 333.33, N = 3 137000 129667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 333.33, N = 3 137000 129667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 1.53, N = 3 SE +/- 2.31, N = 3 2139 2026 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 44.28, N = 3 SE +/- 34.36, N = 3 3156 3330 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 33.33, N = 3 SE +/- 115.47, N = 3 10333 10900 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 2 Threads OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 2 Threads 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.07, N = 3 SE +/- 0.20, N = 3 76.71 80.85 1. (CC) gcc options: -O2 -static -pthread -lrt
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 0.58, N = 3 SE +/- 3.76, N = 3 2134 2025 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 0.88, N = 3 SE +/- 2.03, N = 3 2416 2293 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 17K 34K 51K 68K 85K SE +/- 88.19, N = 3 77300 73367 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.15, N = 3 SE +/- 0.67, N = 3 2581 2451 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.73, N = 3 2762 2623 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 762.31, N = 3 SE +/- 808.98, N = 3 77433 81533 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 27.68, N = 4 SE +/- 1.20, N = 3 1760 1672 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50K 100K 150K 200K 250K SE +/- 3473.11, N = 4 225250 214000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.20, N = 3 2594 2465 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
RAMspeed SMP Type: Average - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Average - Benchmark: Integer 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K 20288.79 21327.26 1. (CC) gcc options: -O3 -march=native
Tensorflow Build: Cifar10 OpenBenchmarking.org Seconds, Fewer Is Better Tensorflow 2017-02-03 Build: Cifar10 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 8 16 24 32 40 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 34.47 32.82
RAMspeed SMP Type: Copy - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Integer 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4K 8K 12K 16K 20K 19682.27 18773.03 1. (CC) gcc options: -O3 -march=native
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.45, N = 3 SE +/- 5.46, N = 3 2620 2499 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 14.75, N = 3 SE +/- 25.46, N = 3 2431 2548 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.73, N = 3 SE +/- 18.48, N = 3 2523 2642 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.45, N = 3 SE +/- 13.86, N = 3 2491 2607 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.60, N = 3 SE +/- 19.34, N = 3 2463 2576 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 9K 18K 27K 36K 45K SE +/- 57.74, N = 3 SE +/- 288.68, N = 3 39400 41200 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Dbench Client Count: 256 OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 256 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 14.29, N = 3 SE +/- 17.35, N = 3 8087.02 7736.44 1. (CC) gcc options: -lpopt -O2
FS-Mark Test: 5000 Files, 1MB Size, 4 Threads OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 5000 Files, 1MB Size, 4 Threads 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 5.46, N = 3 SE +/- 12.84, N = 3 1516.80 1585.47 1. (CC) gcc options: -static
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 3.21, N = 3 SE +/- 1.76, N = 3 2508 2402 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 12.72, N = 3 SE +/- 5.24, N = 3 1809 1736 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 2.19, N = 3 SE +/- 4.06, N = 3 2520 2419 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.19, N = 3 SE +/- 7.06, N = 3 2656 2551 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50K 100K 150K 200K 250K SE +/- 1452.97, N = 3 SE +/- 881.92, N = 3 231333 222333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50K 100K 150K 200K 250K SE +/- 3844.19, N = 3 227667 219000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 29.48, N = 3 SE +/- 1.33, N = 3 1778 1713 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
POV-Ray Trace Time OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6 12 18 24 30 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 27.62 26.63 -lSDL 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 5.29, N = 3 SE +/- 23.10, N = 3 2565 2660 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 8K 16K 24K 32K 40K SE +/- 33.33, N = 3 SE +/- 33.33, N = 3 38733 37433 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 8K 16K 24K 32K 40K 38800 37500 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 0.88, N = 3 SE +/- 1.15, N = 3 2420 2340 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 1.15, N = 3 2424 2345 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Go Benchmarks Test: http OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: http 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 58.71, N = 3 SE +/- 109.33, N = 3 8031 8297
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 2.96, N = 3 SE +/- 18.77, N = 3 2572 2651 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 333.33, N = 3 129667 126000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 5.20, N = 3 2024 1969 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 3.71, N = 3 SE +/- 14.17, N = 3 2684 2614 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 0.88, N = 3 2289 2234 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
SQLite Timed SQLite Insertions OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.22 Timed SQLite Insertions 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.477 0.954 1.431 1.908 2.385 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 2.07 2.12 1. (CC) gcc options: -O2 -lz -ldl -lpthread
RAMspeed SMP Type: Add - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Add - Benchmark: Floating Point 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K 21799.18 22325.64 1. (CC) gcc options: -O3 -march=native
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 1.20, N = 3 SE +/- 0.67, N = 3 2297 2243 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Stream Type: Copy OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 374.79, N = 5 SE +/- 58.42, N = 5 81634.76 79721.98 1. (CC) gcc options: -O3 -march=native -fopenmp
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 16K 32K 48K 64K 80K SE +/- 33.33, N = 3 SE +/- 33.33, N = 3 73467 71767 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 16K 32K 48K 64K 80K SE +/- 33.33, N = 3 73200 71533 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 0.58, N = 3 SE +/- 9.33, N = 3 2493 2439 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Multichase Pointer Chaser Test: 4MB Array, 64 Byte Stride OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 4MB Array, 64 Byte Stride 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 9.50 9.70 1. (CC) gcc options: -O2 -static -pthread -lrt
RAMspeed SMP Type: Triad - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Triad - Benchmark: Floating Point 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K 21730.75 21301.65 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Copy - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Copy - Benchmark: Floating Point 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4K 8K 12K 16K 20K 19182.32 18838.58 1. (CC) gcc options: -O3 -march=native
Multichase Pointer Chaser Test: 256MB Array, 256 Byte Stride OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 256MB Array, 256 Byte Stride 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 16 32 48 64 80 SE +/- 1.20, N = 4 SE +/- 0.25, N = 3 71.57 72.86 1. (CC) gcc options: -O2 -static -pthread -lrt
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 7.00, N = 3 SE +/- 10.41, N = 3 2685 2732 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 3.93, N = 3 SE +/- 12.35, N = 3 2577 2537 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
RAMspeed SMP Type: Scale - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Scale - Benchmark: Floating Point 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4K 8K 12K 16K 20K 19611.31 19834.23 1. (CC) gcc options: -O3 -march=native
Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 4 Threads OpenBenchmarking.org ns, Fewer Is Better Multichase Pointer Chaser Test: 1GB Array, 256 Byte Stride, 4 Threads 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.39, N = 3 82.12 83.02 1. (CC) gcc options: -O2 -static -pthread -lrt
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 10.58, N = 3 SE +/- 1.00, N = 3 1594 1611 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 1527.53, N = 3 204000 206000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 34.08, N = 3 SE +/- 41.60, N = 3 3066 3093 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 1.45, N = 3 SE +/- 12.41, N = 3 2615 2593 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
OSBench Test: Launch Programs OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11 22 33 44 55 SE +/- 0.23, N = 3 SE +/- 0.30, N = 3 50.15 49.74 1. (CC) gcc options: -lm
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 26.58, N = 3 SE +/- 32.91, N = 3 2921 2910 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Y-Cruncher Calculating 500M Pi Digits OpenBenchmarking.org Seconds, Fewer Is Better Y-Cruncher 0.7.5.9481 Calculating 500M Pi Digits 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 16.29 16.23
RAMspeed SMP Type: Average - Benchmark: Floating Point OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Average - Benchmark: Floating Point 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4K 8K 12K 16K 20K 20567.55 20641.24 1. (CC) gcc options: -O3 -march=native
RAMspeed SMP Type: Add - Benchmark: Integer OpenBenchmarking.org MB/s, More Is Better RAMspeed SMP 3.5.0 Type: Add - Benchmark: Integer 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K 21596.76 21564.59 1. (CC) gcc options: -O3 -march=native
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 11.32, N = 3 SE +/- 18.49, N = 3 2736 2734 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K SE +/- 217.94, N = 4 12650 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K 12000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K 12100 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 33.33, N = 3 10333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K SE +/- 33.33, N = 3 12233 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 33.33, N = 3 10233 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K SE +/- 284.31, N = 4 12550 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K SE +/- 204.12, N = 4 12400 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K SE +/- 33.33, N = 3 12333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon Gold 6140 3K 6K 9K 12K 15K 12300 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Cyclictest Test Case: clock_nanosleep TIME_ABSTIME, Interval 10000 ms, 10000 Loops OpenBenchmarking.org ms Average, Fewer Is Better Cyclictest 0.84 Test Case: clock_nanosleep TIME_ABSTIME, Interval 10000 ms, 10000 Loops 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.45 0.9 1.35 1.8 2.25 2 2 1. (CC) gcc options: -O2 -lrt -lpthread -lrttest
Hackbench Count: 32 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 32 - Type: Thread 2 x Xeon E5-2690 v4 8 16 24 32 40 SE +/- 0.07, N = 3 33.76 1. (CC) gcc options: -lpthread
FFmpeg H.264 HD To NTSC DV OpenBenchmarking.org Seconds, Fewer Is Better FFmpeg 3.4.1 H.264 HD To NTSC DV 2 x Xeon E5-2690 v4 3 6 9 12 15 SE +/- 0.14, N = 3 9.08 1. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -lXv -lX11 -lXext -lSDL2 -lxcb -lxcb-xfixes -lxcb-shape -lsndio -lasound -lm -ldl -llzma -lbz2 -pthread -std=c11 -fomit-frame-pointer -fPIC -O3 -fno-math-errno -fno-signed-zeros -fno-tree-vectorize -MMD -MF -MT
Parallel BZIP2 Compression 256MB File Compression OpenBenchmarking.org Seconds, Fewer Is Better Parallel BZIP2 Compression 1.1.12 256MB File Compression 2 x Xeon E5-2690 v4 0.549 1.098 1.647 2.196 2.745 SE +/- 0.05, N = 3 2.44 1. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread
VP9 libvpx Encoding vpxenc OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.7.0 vpxenc 2 x Xeon E5-2690 v4 0.2295 0.459 0.6885 0.918 1.1475 SE +/- 0.00, N = 3 1.02 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE
High Performance Conjugate Gradient OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.0 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.2138 0.4276 0.6414 0.8552 1.069 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 0.95 0.95
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 1.86, N = 3 SE +/- 9.56, N = 3 2486 2486 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
OSBench Test: Create Processes OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 10 20 30 40 50 SE +/- 0.47, N = 3 SE +/- 0.99, N = 6 42.43 40.02 1. (CC) gcc options: -lm
Sunflow Rendering System Global Illumination + Image Synthesis OpenBenchmarking.org Seconds, Fewer Is Better Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.2183 0.4366 0.6549 0.8732 1.0915 SE +/- 0.01, N = 3 SE +/- 0.02, N = 6 0.97 0.87
Redis Test: GET OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400K 800K 1200K 1600K 2000K SE +/- 52359.61, N = 6 SE +/- 60376.06, N = 6 1377549.17 2063823.71 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Redis Test: LPOP OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPOP 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500K 1000K 1500K 2000K 2500K SE +/- 48541.62, N = 6 SE +/- 48406.57, N = 6 1518694.56 2265203.02 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
Hackbench Count: 16 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Process 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.20, N = 3 SE +/- 2.01, N = 6 17.94 19.11 1. (CC) gcc options: -lpthread
Hackbench Count: 8 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Process 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 3 6 9 12 15 SE +/- 0.19, N = 6 SE +/- 0.56, N = 6 10.47 10.06 1. (CC) gcc options: -lpthread
Hackbench Count: 4 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Process 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.12, N = 6 SE +/- 0.47, N = 6 5.86 7.49 1. (CC) gcc options: -lpthread
Hackbench Count: 2 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Process 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.1925 2.385 3.5775 4.77 5.9625 SE +/- 0.10, N = 6 SE +/- 0.06, N = 3 3.77 5.30 1. (CC) gcc options: -lpthread
Hackbench Count: 16 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Thread 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.21, N = 3 SE +/- 0.58, N = 6 19.10 18.14 1. (CC) gcc options: -lpthread
Hackbench Count: 1 - Type: Process OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Process 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.7403 1.4806 2.2209 2.9612 3.7015 SE +/- 0.15, N = 6 SE +/- 0.02, N = 3 2.98 3.29 1. (CC) gcc options: -lpthread
Hackbench Count: 4 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Thread 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2 4 6 8 10 SE +/- 0.18, N = 6 SE +/- 0.63, N = 6 6.07 6.92 1. (CC) gcc options: -lpthread
Hackbench Count: 2 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Thread 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1.2195 2.439 3.6585 4.878 6.0975 SE +/- 0.16, N = 6 SE +/- 0.09, N = 3 4.13 5.42 1. (CC) gcc options: -lpthread
Hackbench Count: 1 - Type: Thread OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 1 - Type: Thread 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 0.7358 1.4716 2.2074 2.9432 3.679 SE +/- 0.16, N = 6 SE +/- 0.07, N = 6 3.27 3.17 1. (CC) gcc options: -lpthread
ebizzy OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200K 400K 600K 800K 1000K SE +/- 10481.42, N = 4 SE +/- 24706.52, N = 6 623000 943906 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
Cpuminer-Opt Algorithm: cryptonight OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: cryptonight 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 200 400 600 800 1000 SE +/- 12.31, N = 6 SE +/- 47.07, N = 6 630.06 796.30 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Cpuminer-Opt Algorithm: myr-gr OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: myr-gr 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1300 2600 3900 5200 6500 SE +/- 1977.56, N = 6 SE +/- 0.02, N = 3 6238.62 11.93 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
Cpuminer-Opt Algorithm: deep OpenBenchmarking.org Hash Speed, More Is Better Cpuminer-Opt 3.8.3.3 Algorithm: deep 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5 10 15 20 25 SE +/- 0.23, N = 3 SE +/- 0.99, N = 6 16.02 19.01 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
IOzone Record Size: 64Kb - File Size: 512MB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 512MB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1300 2600 3900 5200 6500 SE +/- 9.76, N = 3 SE +/- 452.68, N = 6 5546.56 6005.59 1. (CC) gcc options: -O3
IOzone Record Size: 1MB - File Size: 512MB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 512MB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 281.49, N = 6 SE +/- 28.96, N = 3 4989.07 6830.55 1. (CC) gcc options: -O3
IOzone Record Size: 64Kb - File Size: 8GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 64Kb - File Size: 8GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 422.96, N = 6 SE +/- 314.80, N = 6 5525.86 7000.87 1. (CC) gcc options: -O3
IOzone Record Size: 4Kb - File Size: 8GB - Disk Test: Write Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 8GB - Disk Test: Write Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 32.27, N = 6 SE +/- 1.92, N = 3 650.55 1201.49 1. (CC) gcc options: -O3
IOzone Record Size: 4Kb - File Size: 8GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 8GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 211.79, N = 6 SE +/- 29.10, N = 3 3798.10 5165.63 1. (CC) gcc options: -O3
IOzone Record Size: 4Kb - File Size: 2GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 4Kb - File Size: 2GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 172.75, N = 6 SE +/- 4.39, N = 3 3234.92 5078.78 1. (CC) gcc options: -O3
IOzone Record Size: 1MB - File Size: 8GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 8GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 309.24, N = 6 SE +/- 274.61, N = 6 4700.61 6607.57 1. (CC) gcc options: -O3
IOzone Record Size: 1MB - File Size: 4GB - Disk Test: Read Performance OpenBenchmarking.org MB/s, More Is Better IOzone 3.465 Record Size: 1MB - File Size: 4GB - Disk Test: Read Performance 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1200 2400 3600 4800 6000 SE +/- 24.98, N = 3 SE +/- 328.42, N = 6 5618.88 5329.01 1. (CC) gcc options: -O3
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 95.19, N = 6 SE +/- 31.10, N = 3 2160 2114 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 5.21, N = 3 SE +/- 223.00, N = 6 2433 2876 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 138.85, N = 6 SE +/- 183.79, N = 6 2467 2971 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 800.42, N = 6 SE +/- 1444.45, N = 6 18900 24233 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 100.45, N = 6 SE +/- 180.90, N = 6 2361 3029 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 136.70, N = 6 SE +/- 204.14, N = 6 2042 2609 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 9K 18K 27K 36K 45K SE +/- 133.33, N = 3 SE +/- 3314.35, N = 6 29333 43483 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 8.02, N = 3 SE +/- 207.07, N = 6 1833 2718 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 107.56, N = 6 SE +/- 6.94, N = 3 2272 2954 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 145.30, N = 3 SE +/- 4931.42, N = 6 75367 79483 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 4.98, N = 3 SE +/- 154.14, N = 6 2355 2484 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 132.09, N = 6 SE +/- 196.03, N = 6 2143 2310 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 3390.35, N = 6 SE +/- 666.67, N = 3 130167 181333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 53.11, N = 6 SE +/- 9.74, N = 3 2038 2831 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 16.50, N = 3 SE +/- 214.39, N = 6 2421 2853 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 864.65, N = 6 SE +/- 1745.02, N = 6 18283 23133 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 108.17, N = 6 SE +/- 222.91, N = 6 2285 2914 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 548.74, N = 3 SE +/- 1474.83, N = 6 16933 24733 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1300 2600 3900 5200 6500 SE +/- 79.49, N = 3 SE +/- 382.51, N = 6 4307 6227 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12K 24K 36K 48K 60K SE +/- 1799.95, N = 6 SE +/- 264.58, N = 3 40450 57700 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 234.00, N = 6 SE +/- 33.24, N = 3 5077 7210 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 105.22, N = 6 SE +/- 0.67, N = 3 2251 2903 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 60K 120K 180K 240K 300K SE +/- 3145.76, N = 4 SE +/- 16787.73, N = 6 186750 274833 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 24.56, N = 4 SE +/- 131.38, N = 6 1458 2146 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 121.87, N = 6 SE +/- 180.03, N = 6 2056 2334 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 90K 180K 270K 360K 450K SE +/- 11999.07, N = 6 SE +/- 16390.38, N = 6 321333 429333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 46.68, N = 6 SE +/- 63.64, N = 6 1255 1677 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 120.10, N = 6 SE +/- 1.15, N = 3 2066 2932 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 128.67, N = 6 SE +/- 174.49, N = 6 2017 2605 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 132.40, N = 6 SE +/- 193.86, N = 6 2199 2640 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 11K 22K 33K 44K 55K SE +/- 1877.66, N = 6 SE +/- 1502.00, N = 6 32783 49500 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 117.76, N = 6 SE +/- 83.82, N = 6 2049 3104 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 138.06, N = 6 SE +/- 22.67, N = 3 2101 2964 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 101.21, N = 6 SE +/- 173.87, N = 6 1937 2378 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 6009.25, N = 6 SE +/- 10515.86, N = 6 126333 149500 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 93.73, N = 6 SE +/- 165.60, N = 6 1975 2337 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 300.00, N = 3 SE +/- 4691.16, N = 6 84300 102183 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 1.73, N = 3 SE +/- 293.65, N = 6 5310 6393 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 5497.98, N = 6 SE +/- 1527.53, N = 3 144833 195000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1300 2600 3900 5200 6500 SE +/- 170.84, N = 6 SE +/- 49.27, N = 3 4525 6092 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6K 12K 18K 24K 30K SE +/- 850.75, N = 6 SE +/- 100.00, N = 3 17933 27600 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 211.79, N = 6 SE +/- 25.11, N = 3 4490 6903 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12K 24K 36K 48K 60K SE +/- 1774.64, N = 6 SE +/- 693.62, N = 3 41700 56867 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 221.43, N = 6 SE +/- 87.11, N = 3 5211 7109 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 50.00, N = 2 SE +/- 404.15, N = 3 10050 10800 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 582.38, N = 6 SE +/- 264.58, N = 3 22550 25100 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 72.82, N = 6 SE +/- 34.28, N = 3 2820 3138 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 103.60, N = 6 SE +/- 191.50, N = 6 1932 2474 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 60K 120K 180K 240K 300K SE +/- 666.67, N = 3 SE +/- 10107.48, N = 6 229333 296833 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 5.33, N = 3 SE +/- 79.59, N = 6 1793 2319 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 132.65, N = 6 SE +/- 179.52, N = 6 2103 2326 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 90K 180K 270K 360K 450K SE +/- 3214.55, N = 3 SE +/- 18481.97, N = 6 345000 433500 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 12.73, N = 3 SE +/- 72.12, N = 6 1348 1693 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 103.95, N = 6 SE +/- 149.82, N = 6 2291 2229 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 121.54, N = 6 SE +/- 198.71, N = 6 2117 2545 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 85.57, N = 6 SE +/- 16.50, N = 3 3329 4518 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 110K 220K 330K 440K 550K SE +/- 9236.40, N = 6 SE +/- 4041.45, N = 3 352333 525000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 72.24, N = 6 SE +/- 30.78, N = 3 2754 4104 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 86.83, N = 6 SE +/- 15.72, N = 3 3280 4465 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 2531.46, N = 6 SE +/- 881.92, N = 3 72117 108667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1500 3000 4500 6000 7500 SE +/- 157.68, N = 6 SE +/- 48.58, N = 3 4507 6783 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 92.40, N = 6 SE +/- 32.06, N = 3 3399 4572 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 5385.16, N = 6 SE +/- 1000.00, N = 3 136000 195000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1300 2600 3900 5200 6500 SE +/- 168.48, N = 6 SE +/- 27.17, N = 3 4250 6093 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 85.06, N = 6 SE +/- 10.41, N = 3 3419 4534 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 70K 140K 210K 280K 350K SE +/- 6505.13, N = 6 246500 328000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 100.25, N = 6 SE +/- 19.03, N = 3 3853 5125 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 2K 4K 6K 8K 10K SE +/- 33.33, N = 3 SE +/- 404.15, N = 3 10367 10800 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 110K 220K 330K 440K 550K SE +/- 10553.83, N = 6 SE +/- 1527.53, N = 3 364500 523000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 82.40, N = 6 SE +/- 12.12, N = 3 2850 4086 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 80.57, N = 6 SE +/- 7.51, N = 3 3192 4475 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 89.26, N = 6 SE +/- 165.60, N = 6 2004 2542 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 100.91, N = 6 SE +/- 201.03, N = 6 2082 2828 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 104.98, N = 6 SE +/- 2.08, N = 3 2234 2357 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 4K 8K 12K 16K 20K SE +/- 33.33, N = 3 SE +/- 1755.50, N = 6 19167 20833 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 5.51, N = 3 SE +/- 220.77, N = 6 2400 2604 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 69.37, N = 6 SE +/- 29.87, N = 3 2827 3149 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 104.26, N = 6 SE +/- 149.40, N = 6 1984 2402 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 132.94, N = 6 SE +/- 171.15, N = 6 2155 2696 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 3084.26, N = 6 65800 92300 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 96.38, N = 6 2056 2885 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 84.88, N = 6 SE +/- 144.56, N = 6 2356 2428 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 333.33, N = 3 SE +/- 10056.51, N = 6 128667 149000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 2.33, N = 3 SE +/- 156.85, N = 6 2011 2327 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 17.33, N = 3 SE +/- 166.05, N = 6 1724 2688 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 109.27, N = 6 SE +/- 223.39, N = 6 2108 2813 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 1005.32, N = 6 SE +/- 1642.85, N = 6 17100 22817 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 125.83, N = 6 SE +/- 217.11, N = 6 2139 2894 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 512KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 82.83, N = 6 SE +/- 12.06, N = 3 3246 4436 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 5K 10K 15K 20K 25K SE +/- 1096.18, N = 6 SE +/- 202.76, N = 3 18983 25233 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 273.70, N = 6 SE +/- 48.55, N = 3 4748 6306 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 12K 24K 36K 48K 60K SE +/- 1328.57, N = 6 SE +/- 266.67, N = 3 42233 53867 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 128KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 166.06, N = 6 SE +/- 35.51, N = 3 5280 6735 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 125.87, N = 6 SE +/- 6.67, N = 3 2218 2851 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50K 100K 150K 200K 250K SE +/- 8880.69, N = 6 SE +/- 14235.32, N = 6 191000 240333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 69.28, N = 6 SE +/- 111.72, N = 6 1492 1879 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 134.65, N = 6 SE +/- 222.56, N = 6 1972 2370 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 90.71, N = 6 SE +/- 103.31, N = 6 2148 2055 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 166.03, N = 6 2486 2719 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 10K 20K 30K 40K 50K SE +/- 1484.91, N = 6 SE +/- 491.03, N = 3 30317 48933 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 93.50, N = 6 1895 3058 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 20K 40K 60K 80K 100K SE +/- 2684.90, N = 6 SE +/- 5492.76, N = 6 67200 84267 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 87.16, N = 6 SE +/- 173.23, N = 6 2110 2639 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 104.57, N = 6 SE +/- 180.03, N = 6 2330 2563 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 5134.31, N = 6 SE +/- 9570.79, N = 6 114833 153000 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 79.60, N = 6 SE +/- 149.66, N = 6 1797 2391 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 16K 32K 48K 64K 80K SE +/- 3428.97, N = 6 SE +/- 1337.08, N = 3 75450 64567 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 64KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 213.77, N = 6 SE +/- 64.50, N = 3 4716 4055 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 83.95, N = 6 SE +/- 15.17, N = 3 3324 4057 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 4912.57, N = 6 SE +/- 7747.76, N = 6 128000 147833 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1000 2000 3000 4000 5000 SE +/- 152.02, N = 6 SE +/- 242.92, N = 6 4007 4615 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40K 80K 120K 160K 200K SE +/- 6052.09, N = 6 SE +/- 333.33, N = 3 171833 209667 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 16KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 96.14, N = 6 SE +/- 4.16, N = 3 2683 3272 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 6K 12K 18K 24K 30K SE +/- 886.85, N = 6 SE +/- 328.30, N = 3 17150 26033 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 256KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1400 2800 4200 5600 7000 SE +/- 221.86, N = 6 SE +/- 81.54, N = 3 4287 6517 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 132.92, N = 6 SE +/- 217.32, N = 6 2086 2354 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 50K 100K 150K 200K 250K SE +/- 333.33, N = 3 SE +/- 16313.08, N = 6 162667 236500 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 8KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 400 800 1200 1600 2000 SE +/- 3.84, N = 3 SE +/- 127.33, N = 6 1273 1849 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 600 1200 1800 2400 3000 SE +/- 131.34, N = 6 SE +/- 117.60, N = 6 2199 2712 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 80K 160K 240K 320K 400K SE +/- 10842.56, N = 6 SE +/- 15334.96, N = 6 279833 354833 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 300 600 900 1200 1500 SE +/- 41.95, N = 6 SE +/- 60.21, N = 6 1094 1386 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 144.78, N = 6 SE +/- 198.76, N = 6 2070 2484 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 500 1000 1500 2000 2500 SE +/- 114.20, N = 6 SE +/- 185.31, N = 6 2068 2466 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 8MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 91.48, N = 6 SE +/- 13.65, N = 3 3317 4017 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 3K 6K 9K 12K 15K SE +/- 1305.54, N = 6 14567 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 13 26 39 52 65 SE +/- 5.09, N = 6 SE +/- 0.03, N = 3 57.00 31.73 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: Yes - Block Size: 2MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 85.81, N = 6 SE +/- 13.04, N = 3 3336 4029 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 30K 60K 90K 120K 150K SE +/- 4996.11, N = 6 SE +/- 6535.37, N = 6 123833 157333 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 32KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 1100 2200 3300 4400 5500 SE +/- 155.59, N = 6 SE +/- 206.35, N = 6 3865 4920 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: No - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 700 1400 2100 2800 3500 SE +/- 48.08, N = 6 SE +/- 1.53, N = 3 1927 3102 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 9K 18K 27K 36K 45K SE +/- 1285.50, N = 6 40650 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 40 80 120 160 200 SE +/- 5.08, N = 6 SE +/- 0.03, N = 3 159.00 31.83 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 1MB - Disk Target: /var 2 x Xeon E5-2690 v4 2 x Xeon Gold 6140 900 1800 2700 3600 4500 SE +/- 89.96, N = 6 SE +/- 24.90, N = 3 3251 4043 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Phoronix Test Suite v10.8.5