AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (2702 BIOS) and Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB on Ubuntu 20.10 via the Phoronix Test Suite.
Linux 5.10.2 Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (2702 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211
OS: Ubuntu 20.10, Kernel: 5.10.2-051002-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009Graphics Notes: GLAMORPython Notes: Python 3.8.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Linux 5.11 Git OS: Ubuntu 20.10, Kernel: 5.10.0-051000daily20201222-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160
HPC Challenge HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-HPL Linux 5.10.2 Linux 5.11 Git 10 20 30 40 50 SE +/- 0.11, N = 3 SE +/- 0.19, N = 3 44.89 45.08 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: Jetstream - Browser: Firefox Linux 5.10.2 Linux 5.11 Git 40 80 120 160 200 SE +/- 0.60, N = 3 SE +/- 1.39, N = 12 179.95 162.75 1. firefox 84.0
VkFFT VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Benchmark Score, More Is Better VkFFT 1.1.1 Linux 5.10.2 Linux 5.11 Git 3K 6K 9K 12K 15K SE +/- 63.74, N = 3 SE +/- 53.28, N = 3 15204 14556 1. (CXX) g++ options: -O3 -pthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 600 1200 1800 2400 3000 SE +/- 24.19, N = 15 SE +/- 37.58, N = 3 2958.33 2916.87 MIN: 2727.04 MIN: 2849.92 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 22.45, N = 3 SE +/- 19.15, N = 15 1813.94 1811.34 MIN: 1762.3 MIN: 1701.55 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 Linux 5.10.2 Linux 5.11 Git 1.1185 2.237 3.3555 4.474 5.5925 SE +/- 0.00425, N = 3 SE +/- 0.00574, N = 3 4.96530 4.97109 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
Unigine Valley This test calculates the average frame-rate within the Valley demo for the Unigine engine, released in February 2013. This engine is extremely demanding on the system's graphics card. Unigine Valley relies upon an OpenGL 3 core profile context. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Unigine Valley 1.0 Resolution: 1920 x 1080 - Mode: Fullscreen - Renderer: OpenGL Linux 5.10.2 Linux 5.11 Git 30 60 90 120 150 SE +/- 0.31, N = 3 SE +/- 1.01, N = 3 153.32 141.16
OpenBenchmarking.org Frames Per Second, More Is Better Unigine Valley 1.0 Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGL Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.06, N = 3 SE +/- 0.45, N = 3 92.64 89.74
Unigine Superposition This test calculates the average frame-rate within the Superposition demo for the Unigine engine, released in 2017. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Unigine Superposition 1.0 Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Ultra - Renderer: OpenGL Linux 5.10.2 Linux 5.11 Git 7 14 21 28 35 SE +/- 0.03, N = 3 28.1 28.0 MAX: 32.9 MAX: 32.7
OpenBenchmarking.org Frames Per Second, More Is Better Unigine Superposition 1.0 Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: High - Renderer: OpenGL Linux 5.10.2 Linux 5.11 Git 16 32 48 64 80 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 70.6 70.5 MAX: 84.7 MAX: 84.7
OpenBenchmarking.org Frames Per Second, More Is Better Unigine Superposition 1.0 Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Low - Renderer: OpenGL Linux 5.10.2 Linux 5.11 Git 50 100 150 200 250 SE +/- 0.58, N = 3 SE +/- 1.93, N = 3 209.9 202.1 MAX: 314.8 MAX: 317.9
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: CanvasMark - Browser: Firefox Linux 5.10.2 Linux 5.11 Git 3K 6K 9K 12K 15K SE +/- 50.80, N = 3 SE +/- 126.10, N = 3 16016 15653 1. firefox 84.0
iPerf iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 10 Linux 5.10.2 Linux 5.11 Git 20K 40K 60K 80K 100K SE +/- 1106.82, N = 15 SE +/- 691.66, N = 12 102794 96016 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: Ultimate Linux 5.10.2 Linux 5.11 Git 70 140 210 280 350 SE +/- 0.42, N = 3 SE +/- 4.77, N = 15 329.77 239.83 MIN: 63 / MAX: 541 MIN: 56 / MAX: 543
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 300 600 900 1200 1500 SE +/- 183.92, N = 12 SE +/- 138.35, N = 15 1569 1311 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 700 1400 2100 2800 3500 SE +/- 367.90, N = 12 SE +/- 276.68, N = 15 3145 2629 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 118.43, N = 15 SE +/- 129.36, N = 12 1931 1952 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 800 1600 2400 3200 4000 SE +/- 236.96, N = 15 SE +/- 258.72, N = 12 3869 3911 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org VKMark Score, More Is Better VKMark 2020-05-21 Resolution: 2560 x 1440 Linux 5.10.2 Linux 5.11 Git 1000 2000 3000 4000 5000 SE +/- 3.38, N = 3 SE +/- 2.73, N = 3 4573 4508 1. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.org VKMark Score, More Is Better VKMark 2020-05-21 Resolution: 3840 x 2160 Linux 5.10.2 Linux 5.11 Git 500 1000 1500 2000 2500 SE +/- 0.67, N = 3 SE +/- 3.51, N = 3 2275 2259 1. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
Ethr Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds, Fewer Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1 Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.36, N = 12 SE +/- 0.20, N = 15 17.47 18.00 MIN: 6.25 / MAX: 24.09 MIN: 7.07 / MAX: 24.26
OpenBenchmarking.org Connections/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 16 Linux 5.10.2 Linux 5.11 Git 20K 40K 60K 80K 100K SE +/- 11316.74, N = 13 SE +/- 10244.72, N = 15 73000 102341
OpenBenchmarking.org Connections/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 8 Linux 5.10.2 Linux 5.11 Git 12K 24K 36K 48K 60K SE +/- 1144.33, N = 12 SE +/- 1166.03, N = 15 55999 57724
InfluxDB This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Linux 5.10.2 Linux 5.11 Git 200K 400K 600K 800K 1000K SE +/- 2020.53, N = 3 SE +/- 451.41, N = 3 1074910.6 1033610.1
iPerf iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 1 Linux 5.10.2 Linux 5.11 Git 20K 40K 60K 80K 100K SE +/- 1250.48, N = 3 SE +/- 1040.04, N = 15 92774 90018 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
Ethr Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds, Fewer Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 16 Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.21, N = 4 SE +/- 0.15, N = 15 16.72 17.60 MIN: 6.95 / MAX: 21.64 MIN: 6.12 / MAX: 24.22
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average Latency Linux 5.10.2 Linux 5.11 Git 0.0097 0.0194 0.0291 0.0388 0.0485 SE +/- 0.001, N = 3 SE +/- 0.003, N = 15 0.032 0.043 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 1 - Mode: Read Only Linux 5.10.2 Linux 5.11 Git 7K 14K 21K 28K 35K SE +/- 302.18, N = 3 SE +/- 1640.45, N = 15 31316 24534 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 70K 140K 210K 280K 350K SE +/- 2645.75, N = 3 SE +/- 3343.03, N = 15 328000 328067 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 300 600 900 1200 1500 SE +/- 11.61, N = 3 SE +/- 13.03, N = 15 1281 1281 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: High Linux 5.10.2 Linux 5.11 Git 100 200 300 400 500 SE +/- 0.74, N = 3 SE +/- 25.17, N = 12 471.83 328.09 MIN: 190 / MAX: 698 MIN: 165 / MAX: 697
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 160K 320K 480K 640K 800K SE +/- 8089.77, N = 3 SE +/- 6182.05, N = 15 717333 736467 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 600 1200 1800 2400 3000 SE +/- 32.59, N = 3 SE +/- 24.12, N = 15 2802 2877 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
ET: Legacy ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better ET: Legacy 2.75 Renderer: Renderer2 - Resolution: 3840 x 2160 Linux 5.10.2 Linux 5.11 Git 60 120 180 240 300 SE +/- 1.23, N = 3 SE +/- 8.51, N = 15 278.4 182.0
Ethr Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds, Fewer Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 8 Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.19, N = 15 SE +/- 0.18, N = 3 17.61 17.31 MIN: 6.06 / MAX: 24.09 MIN: 8.23 / MAX: 24.16
RealSR-NCNN RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: Yes Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 80.05 79.94
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 100K 200K 300K 400K 500K SE +/- 7295.18, N = 12 SE +/- 666.67, N = 3 442500 460333 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 28.55, N = 12 SE +/- 3.18, N = 3 1729 1799 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenVKL OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Items / Sec, More Is Better OpenVKL 0.9 Benchmark: vklBenchmark Linux 5.10.2 Linux 5.11 Git 50 100 150 200 250 SE +/- 0.59, N = 3 SE +/- 0.97, N = 3 236.14 217.39 MIN: 1 / MAX: 931 MIN: 1 / MAX: 929
Ethr Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 16 Linux 5.10.2 Linux 5.11 Git 40K 80K 120K 160K 200K SE +/- 2208.72, N = 3 SE +/- 1344.75, N = 15 166055.11 164565.42 MIN: 160160 / MAX: 174000 MIN: 151060 / MAX: 174580
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 8 Linux 5.10.2 Linux 5.11 Git 30K 60K 90K 120K 150K SE +/- 1368.78, N = 3 SE +/- 1861.95, N = 15 147520.67 134820.31 MIN: 126280 / MAX: 158780 MIN: 94320 / MAX: 163420
OpenBenchmarking.org Connections/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1 Linux 5.10.2 Linux 5.11 Git 2K 4K 6K 8K 10K SE +/- 58.15, N = 15 SE +/- 70.00, N = 3 7805 7360
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 120K 240K 360K 480K 600K SE +/- 4112.68, N = 12 SE +/- 5925.46, N = 3 580667 570333 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 500 1000 1500 2000 2500 SE +/- 15.89, N = 12 SE +/- 22.50, N = 3 2268 2228 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average Latency Linux 5.10.2 Linux 5.11 Git 0.0081 0.0162 0.0243 0.0324 0.0405 SE +/- 0.000, N = 3 SE +/- 0.002, N = 15 0.028 0.036 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Only Linux 5.10.2 Linux 5.11 Git 8K 16K 24K 32K 40K SE +/- 487.24, N = 3 SE +/- 1834.50, N = 15 36550 29331 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 150K 300K 450K 600K 750K SE +/- 42763.95, N = 12 SE +/- 8569.97, N = 3 714167 705333 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 600 1200 1800 2400 3000 SE +/- 167.02, N = 12 SE +/- 33.26, N = 3 2790 2756 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 9 - Compression Speed Linux 5.10.2 Linux 5.11 Git 15 30 45 60 75 SE +/- 0.22, N = 3 SE +/- 0.68, N = 6 67.06 68.34 1. (CC) gcc options: -O3
KeyDB A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.0.16 Linux 5.10.2 Linux 5.11 Git 170K 340K 510K 680K 850K SE +/- 2699.75, N = 3 SE +/- 1669.87, N = 3 771374.86 752819.27 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms Linux 5.10.2 Linux 5.11 Git 0.2903 0.5806 0.8709 1.1612 1.4515 SE +/- 0.00245, N = 3 SE +/- 0.00139, N = 3 1.29036 1.28612
InfluxDB This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org val/sec, More Is Better InfluxDB 1.8.2 Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000 Linux 5.10.2 Linux 5.11 Git 400K 800K 1200K 1600K 2000K SE +/- 3810.04, N = 3 SE +/- 21384.90, N = 3 1711157.6 1601473.2
Node.js V8 Web Tooling Benchmark Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org runs/s, More Is Better Node.js V8 Web Tooling Benchmark Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.08, N = 3 SE +/- 0.06, N = 3 14.64 15.05 1. Nodejs
v12.18.2
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Sync Linux 5.10.2 Linux 5.11 Git 3K 6K 9K 12K 15K SE +/- 9.81, N = 3 SE +/- 10.35, N = 3 12826 13061 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
CLOMP CLOMP is the C version of the Livermore OpenMP benchmark developed to measure OpenMP overheads and other performance impacts due to threading in order to influence future system designs. This particular test profile configuration is currently set to look at the OpenMP static schedule speed-up across all available CPU cores using the recommended test configuration. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Speedup, More Is Better CLOMP 1.2 Static OMP Speedup Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.28, N = 15 SE +/- 0.10, N = 3 17.2 17.7 1. (CC) gcc options: -fopenmp -O3 -lm
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 120K 240K 360K 480K 600K SE +/- 4430.88, N = 9 SE +/- 7055.34, N = 3 539778 540333 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 500 1000 1500 2000 2500 SE +/- 17.30, N = 9 SE +/- 27.28, N = 3 2109 2111 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Linux 5.10.2 Linux 5.11 Git 300K 600K 900K 1200K 1500K SE +/- 16445.33, N = 3 SE +/- 14903.56, N = 3 1284512 1311891 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Read While Writing Linux 5.10.2 Linux 5.11 Git 800K 1600K 2400K 3200K 4000K SE +/- 49577.80, N = 3 SE +/- 10900.81, N = 3 3582470 3503352 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Read Linux 5.10.2 Linux 5.11 Git 20M 40M 60M 80M 100M SE +/- 126364.22, N = 3 SE +/- 72529.36, N = 3 106394483 105250174 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 3 - Compression Speed Linux 5.10.2 Linux 5.11 Git 16 32 48 64 80 SE +/- 0.76, N = 3 SE +/- 0.86, N = 4 69.64 71.01 1. (CC) gcc options: -O3
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: LargeRandom Linux 5.10.2 Linux 5.11 Git 0.1328 0.2656 0.3984 0.5312 0.664 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 0.59 0.59 1. (CXX) g++ options: -O3 -pthread
NCNN NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: regnety_400m Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 17.33 17.13 MIN: 17.02 / MAX: 18.75 MIN: 16.96 / MAX: 17.68 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: squeezenet_ssd Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 14.20 14.16 MIN: 13.95 / MAX: 15.08 MIN: 13.96 / MAX: 14.52 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: yolov4-tiny Linux 5.10.2 Linux 5.11 Git 5 10 15 20 25 SE +/- 0.09, N = 3 SE +/- 0.46, N = 3 20.07 20.35 MIN: 19.83 / MAX: 20.51 MIN: 19.73 / MAX: 23.56 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: resnet50 Linux 5.10.2 Linux 5.11 Git 6 12 18 24 30 SE +/- 0.05, N = 3 SE +/- 0.33, N = 3 23.80 24.18 MIN: 23.54 / MAX: 31.91 MIN: 23.59 / MAX: 25.19 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: alexnet Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.11, N = 3 SE +/- 0.01, N = 3 10.45 10.34 MIN: 10.28 / MAX: 12.35 MIN: 10.26 / MAX: 10.55 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: resnet18 Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 13.05 13.04 MIN: 12.95 / MAX: 14.92 MIN: 12.96 / MAX: 13.24 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: vgg16 Linux 5.10.2 Linux 5.11 Git 12 24 36 48 60 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 53.43 53.53 MIN: 52.81 / MAX: 60.42 MIN: 52.97 / MAX: 54.2 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: googlenet Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.07, N = 3 SE +/- 0.00, N = 3 12.24 12.15 MIN: 12.06 / MAX: 12.58 MIN: 12.04 / MAX: 12.55 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: blazeface Linux 5.10.2 Linux 5.11 Git 0.414 0.828 1.242 1.656 2.07 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 1.83 1.84 MIN: 1.78 / MAX: 2.05 MIN: 1.77 / MAX: 2.08 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: efficientnet-b0 Linux 5.10.2 Linux 5.11 Git 1.1903 2.3806 3.5709 4.7612 5.9515 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 5.29 5.28 MIN: 5.24 / MAX: 6.68 MIN: 5.24 / MAX: 5.66 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: mnasnet Linux 5.10.2 Linux 5.11 Git 0.8573 1.7146 2.5719 3.4292 4.2865 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 3.81 3.79 MIN: 3.76 / MAX: 4.55 MIN: 3.76 / MAX: 4.2 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: shufflenet-v2 Linux 5.10.2 Linux 5.11 Git 1.0868 2.1736 3.2604 4.3472 5.434 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 4.83 4.81 MIN: 4.78 / MAX: 5.77 MIN: 4.77 / MAX: 4.96 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3 Linux 5.10.2 Linux 5.11 Git 0.9 1.8 2.7 3.6 4.5 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 4.00 3.99 MIN: 3.95 / MAX: 5.33 MIN: 3.95 / MAX: 6.5 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2 Linux 5.10.2 Linux 5.11 Git 0.9495 1.899 2.8485 3.798 4.7475 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 4.22 4.22 MIN: 4.17 / MAX: 5.41 MIN: 4.16 / MAX: 12.09 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: Vulkan GPU - Model: mobilenet Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 12.30 12.20 MIN: 12.12 / MAX: 12.65 MIN: 12.03 / MAX: 14.49 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: regnety_400m Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.36, N = 3 SE +/- 0.07, N = 3 16.68 17.12 MIN: 15.85 / MAX: 18.66 MIN: 16.91 / MAX: 19.11 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: squeezenet_ssd Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 13.98 14.19 MIN: 13.73 / MAX: 14.36 MIN: 13.96 / MAX: 14.58 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: yolov4-tiny Linux 5.10.2 Linux 5.11 Git 5 10 15 20 25 SE +/- 0.10, N = 3 SE +/- 0.48, N = 3 19.89 20.43 MIN: 19.59 / MAX: 20.33 MIN: 19.79 / MAX: 21.71 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet50 Linux 5.10.2 Linux 5.11 Git 6 12 18 24 30 SE +/- 0.10, N = 3 SE +/- 0.36, N = 3 23.62 24.15 MIN: 23.36 / MAX: 24.31 MIN: 23.61 / MAX: 25.38 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: alexnet Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 10.53 10.44 MIN: 10.3 / MAX: 18.31 MIN: 10.28 / MAX: 10.71 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: resnet18 Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.19, N = 3 12.95 13.26 MIN: 12.82 / MAX: 13.21 MIN: 12.97 / MAX: 22.11 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: vgg16 Linux 5.10.2 Linux 5.11 Git 12 24 36 48 60 SE +/- 0.20, N = 3 SE +/- 0.15, N = 3 53.24 53.66 MIN: 52.36 / MAX: 60.28 MIN: 52.97 / MAX: 54.65 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: googlenet Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.30, N = 3 12.22 12.50 MIN: 12.07 / MAX: 13.83 MIN: 12.07 / MAX: 20.49 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: blazeface Linux 5.10.2 Linux 5.11 Git 0.405 0.81 1.215 1.62 2.025 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 1.78 1.80 MIN: 1.73 / MAX: 1.92 MIN: 1.78 / MAX: 2.17 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: efficientnet-b0 Linux 5.10.2 Linux 5.11 Git 1.1948 2.3896 3.5844 4.7792 5.974 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 5.28 5.31 MIN: 5.2 / MAX: 6.76 MIN: 5.24 / MAX: 5.58 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mnasnet Linux 5.10.2 Linux 5.11 Git 0.864 1.728 2.592 3.456 4.32 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 3.78 3.84 MIN: 3.72 / MAX: 4.89 MIN: 3.76 / MAX: 4.16 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: shufflenet-v2 Linux 5.10.2 Linux 5.11 Git 1.0845 2.169 3.2535 4.338 5.4225 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 4.79 4.82 MIN: 4.7 / MAX: 5.85 MIN: 4.78 / MAX: 5.02 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v3-v3 - Model: mobilenet-v3 Linux 5.10.2 Linux 5.11 Git 0.9045 1.809 2.7135 3.618 4.5225 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 3.96 4.02 MIN: 3.9 / MAX: 4.77 MIN: 3.96 / MAX: 4.25 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU-v2-v2 - Model: mobilenet-v2 Linux 5.10.2 Linux 5.11 Git 0.9563 1.9126 2.8689 3.8252 4.7815 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 4.20 4.25 MIN: 4.1 / MAX: 4.92 MIN: 4.16 / MAX: 6.05 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.org ms, Fewer Is Better NCNN 20201218 Target: CPU - Model: mobilenet Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 12.18 12.22 MIN: 12.01 / MAX: 12.69 MIN: 12.05 / MAX: 12.55 1. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GB/s, More Is Better simdjson 0.7.1 Throughput Test: DistinctUserID Linux 5.10.2 Linux 5.11 Git 0.198 0.396 0.594 0.792 0.99 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 0.88 0.88 1. (CXX) g++ options: -O3 -pthread
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Seek Random Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.38, N = 15 14.10 16.43 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org Geometric Mean, More Is Better Selenium Benchmark: Octane - Browser: Firefox Linux 5.10.2 Linux 5.11 Git 5K 10K 15K 20K 25K SE +/- 109.36, N = 3 SE +/- 151.27, N = 3 24398 23912 1. firefox 84.0
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Sequential Fill Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.21, N = 3 SE +/- 0.54, N = 3 81.23 79.68 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Sequential Fill Linux 5.10.2 Linux 5.11 Git 8 16 24 32 40 SE +/- 0.07, N = 3 SE +/- 0.22, N = 3 32.7 33.3 1. (CXX) g++ options: -O3 -lsnappy -lpthread
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: Ultra Linux 5.10.2 Linux 5.11 Git 90 180 270 360 450 SE +/- 1.53, N = 3 SE +/- 0.93, N = 3 411.44 242.93 MIN: 181 / MAX: 618 MIN: 152 / MAX: 555
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Delete Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.19, N = 3 SE +/- 0.55, N = 3 77.71 75.99 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Read Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.062, N = 3 SE +/- 0.342, N = 15 9.996 12.008 1. (CXX) g++ options: -O3 -lsnappy -lpthread
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Hot Read Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.015, N = 3 SE +/- 0.243, N = 15 9.875 11.675 1. (CXX) g++ options: -O3 -lsnappy -lpthread
iPerf iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 5 Linux 5.10.2 Linux 5.11 Git 20K 40K 60K 80K 100K SE +/- 580.37, N = 3 SE +/- 207.53, N = 3 95418 93627 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 10 Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 10.5 10.5 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 5 Linux 5.10.2 Linux 5.11 Git 1.179 2.358 3.537 4.716 5.895 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 5.24 5.24 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.org Mbits/sec, More Is Better iPerf 3.7 Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 1 Linux 5.10.2 Linux 5.11 Git 0.2363 0.4726 0.7089 0.9452 1.1815 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.05 1.05 1. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.org MB/s, More Is Better LZ4 Compression 1.9.3 Compression Level: 1 - Compression Speed Linux 5.10.2 Linux 5.11 Git 3K 6K 9K 12K 15K SE +/- 43.27, N = 3 SE +/- 10.13, N = 3 12158.25 12151.51 1. (CC) gcc options: -O3
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memcpy 1MB Linux 5.10.2 Linux 5.11 Git 8 16 24 32 40 SE +/- 1.54, N = 12 SE +/- 0.20, N = 3 32.70 22.90 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average Latency Linux 5.10.2 Linux 5.11 Git 0.1465 0.293 0.4395 0.586 0.7325 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 0.651 0.647 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 1 - Mode: Read Write Linux 5.10.2 Linux 5.11 Git 300 600 900 1200 1500 SE +/- 8.68, N = 3 SE +/- 4.81, N = 3 1537 1546 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average Latency Linux 5.10.2 Linux 5.11 Git 0.374 0.748 1.122 1.496 1.87 SE +/- 0.002, N = 3 SE +/- 0.003, N = 3 1.662 1.552 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 50 - Mode: Read Write Linux 5.10.2 Linux 5.11 Git 7K 14K 21K 28K 35K SE +/- 34.41, N = 3 SE +/- 52.46, N = 3 30095 32218 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Epoll Wait Linux 5.10.2 Linux 5.11 Git 15K 30K 45K 60K 75K SE +/- 463.05, N = 3 SE +/- 101.53, N = 3 65888 71563 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average Latency Linux 5.10.2 Linux 5.11 Git 0.0288 0.0576 0.0864 0.1152 0.144 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.128 0.127 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 100 - Clients: 50 - Mode: Read Only Linux 5.10.2 Linux 5.11 Git 80K 160K 240K 320K 400K SE +/- 765.65, N = 3 SE +/- 411.53, N = 3 390751 393333 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 500 1000 1500 2000 2500 SE +/- 0.33, N = 3 SE +/- 14.84, N = 3 2380 2357 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 1000 2000 3000 4000 5000 SE +/- 0.88, N = 3 SE +/- 29.94, N = 3 4768 4722 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 100K 200K 300K 400K 500K SE +/- 666.67, N = 3 SE +/- 1527.53, N = 3 476333 473000 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 2.40, N = 3 SE +/- 5.90, N = 3 1861 1847 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 500 1000 1500 2000 2500 SE +/- 0.67, N = 3 2380 2381 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 1000 2000 3000 4000 5000 SE +/- 1.00, N = 3 4767 4770 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 2038 2040 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 900 1800 2700 3600 4500 SE +/- 2.33, N = 3 SE +/- 0.58, N = 3 4083 4086 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 2039 2038 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 900 1800 2700 3600 4500 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 4086 4083 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 0.67, N = 3 SE +/- 0.67, N = 3 2041 2039 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 900 1800 2700 3600 4500 SE +/- 1.86, N = 3 SE +/- 1.33, N = 3 4090 4084 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 0.67, N = 3 2038 2039 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 900 1800 2700 3600 4500 4084 4086 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 120K 240K 360K 480K 600K SE +/- 6806.86, N = 3 SE +/- 4358.90, N = 3 554000 536000 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.10.2 Linux 5.11 Git 500 1000 1500 2000 2500 SE +/- 26.42, N = 3 SE +/- 16.90, N = 3 2164 2095 1. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: System V Message Passing Linux 5.10.2 Linux 5.11 Git 4M 8M 12M 16M 20M SE +/- 80454.79, N = 3 SE +/- 196651.44, N = 3 14501717.00 17617104.63 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Memory Copying Linux 5.10.2 Linux 5.11 Git 300 600 900 1200 1500 SE +/- 1.83, N = 3 SE +/- 2.40, N = 3 1611.73 1615.60 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Malloc Linux 5.10.2 Linux 5.11 Git 40M 80M 120M 160M 200M SE +/- 551101.63, N = 3 SE +/- 266217.26, N = 3 205233129.84 204569594.71 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Matrix Math Linux 5.10.2 Linux 5.11 Git 13K 26K 39K 52K 65K SE +/- 232.35, N = 3 SE +/- 164.05, N = 3 61027.31 60672.25 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: SENDFILE Linux 5.10.2 Linux 5.11 Git 70K 140K 210K 280K 350K SE +/- 318.35, N = 3 SE +/- 274.26, N = 3 289659.24 330425.51 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Atomic Linux 5.10.2 Linux 5.11 Git 40K 80K 120K 160K 200K SE +/- 513.23, N = 3 SE +/- 191.87, N = 3 202668.16 202613.61 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: MMAP Linux 5.10.2 Linux 5.11 Git 80 160 240 320 400 SE +/- 1.25, N = 3 SE +/- 0.38, N = 3 374.66 373.89 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Glibc C String Functions Linux 5.10.2 Linux 5.11 Git 400K 800K 1200K 1600K 2000K SE +/- 12611.55, N = 3 SE +/- 18853.69, N = 3 2076624.77 2053352.36 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Context Switching Linux 5.10.2 Linux 5.11 Git 1.5M 3M 4.5M 6M 7.5M SE +/- 92170.24, N = 3 SE +/- 50290.98, N = 3 6869871.47 6705122.34 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Crypto Linux 5.10.2 Linux 5.11 Git 800 1600 2400 3200 4000 SE +/- 8.27, N = 3 SE +/- 8.97, N = 3 3862.63 3867.49 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: MEMFD Linux 5.10.2 Linux 5.11 Git 200 400 600 800 1000 SE +/- 4.30, N = 3 SE +/- 1.70, N = 3 1106.36 1107.95 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Glibc Qsort Data Sorting Linux 5.10.2 Linux 5.11 Git 40 80 120 160 200 SE +/- 0.70, N = 3 SE +/- 0.45, N = 3 203.95 204.00 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Vector Math Linux 5.10.2 Linux 5.11 Git 30K 60K 90K 120K 150K SE +/- 76.71, N = 3 SE +/- 72.42, N = 3 135272.19 135361.90 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Semaphores Linux 5.10.2 Linux 5.11 Git 400K 800K 1200K 1600K 2000K SE +/- 8971.20, N = 3 SE +/- 4444.66, N = 3 1813741.88 1825250.87 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.11.07 Test: Forking Linux 5.10.2 Linux 5.11 Git 20K 40K 60K 80K 100K SE +/- 158.65, N = 3 SE +/- 365.70, N = 3 81547.64 82828.50 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Lock-Pi Linux 5.10.2 Linux 5.11 Git 150 300 450 600 750 SE +/- 2.33, N = 3 679 677 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Futex Hash Linux 5.10.2 Linux 5.11 Git 1.2M 2.4M 3.6M 4.8M 6M SE +/- 5947.55, N = 3 SE +/- 3392.36, N = 3 5439460 5436582 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.9.0 Binary: Pathtracer ISPC - Model: Asian Dragon Linux 5.10.2 Linux 5.11 Git 5 10 15 20 25 SE +/- 0.19, N = 3 SE +/- 0.16, N = 3 20.73 20.81 MIN: 20.25 / MAX: 21.26 MIN: 20.5 / MAX: 21.49
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Sched Pipe Linux 5.10.2 Linux 5.11 Git 40K 80K 120K 160K 200K SE +/- 2051.49, N = 3 SE +/- 1147.85, N = 3 173668 171415 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
MPV MPV is an open-source, cross-platform media player. This test profile tests the frame-rate that can be achieved unsynchronized in a desynchronized mode. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better MPV Video Input: Big Buck Bunny Sunflower 4K - Decode: Software Only Linux 5.10.2 Linux 5.11 Git 150 300 450 600 750 SE +/- 4.75, N = 3 SE +/- 2.05, N = 3 712.29 628.85 MIN: 444.44 / MAX: 923.08 MIN: 387.1 / MAX: 857.14 1. mpv 0.32.0
Ethr Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 8 Linux 5.10.2 Linux 5.11 Git 900 1800 2700 3600 4500 SE +/- 11.76, N = 3 SE +/- 6.23, N = 3 3969.50 3194.21 MIN: 3740 / MAX: 4100 MIN: 3130 / MAX: 3290
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1 Linux 5.10.2 Linux 5.11 Git 200 400 600 800 1000 SE +/- 0.35, N = 3 SE +/- 1.39, N = 3 1118.25 1125.79 MIN: 1100 / MAX: 1140 MIN: 1100 / MAX: 1150
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 16 Linux 5.10.2 Linux 5.11 Git 1200 2400 3600 4800 6000 SE +/- 6.59, N = 3 SE +/- 4.72, N = 3 4906.65 5622.11 MIN: 4680 / MAX: 5050 MIN: 5440 / MAX: 5760
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 16 Linux 5.10.2 Linux 5.11 Git 50K 100K 150K 200K 250K SE +/- 455.90, N = 3 SE +/- 226.58, N = 3 229572.00 227201.78 MIN: 226350 / MAX: 233820 MIN: 224930 / MAX: 234160
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 8 Linux 5.10.2 Linux 5.11 Git 50K 100K 150K 200K 250K SE +/- 466.03, N = 3 SE +/- 792.15, N = 3 214200.22 214357.78 MIN: 209060 / MAX: 226750 MIN: 208730 / MAX: 228900
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average Latency Linux 5.10.2 Linux 5.11 Git 6 12 18 24 30 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 27.58 27.38 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Write Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 0.90, N = 3 SE +/- 1.95, N = 3 1813 1826 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average Latency Linux 5.10.2 Linux 5.11 Git 0.1397 0.2794 0.4191 0.5588 0.6985 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.621 0.616 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 1 - Mode: Read Write Linux 5.10.2 Linux 5.11 Git 300 600 900 1200 1500 SE +/- 3.56, N = 3 SE +/- 1.80, N = 3 1610 1623 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average Latency Linux 5.10.2 Linux 5.11 Git 0.0214 0.0428 0.0642 0.0856 0.107 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 0.095 0.095 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 13.0 Scaling Factor: 1 - Clients: 50 - Mode: Read Only Linux 5.10.2 Linux 5.11 Git 110K 220K 330K 440K 550K SE +/- 3909.23, N = 3 SE +/- 1197.96, N = 3 524840 524215 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill Linux 5.10.2 Linux 5.11 Git 300K 600K 900K 1200K 1500K SE +/- 5840.17, N = 3 SE +/- 14455.88, N = 6 1408117 1441802 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes Linux 5.10.2 Linux 5.11 Git 6 12 18 24 30 SE +/- 1.45, N = 12 SE +/- 0.88, N = 15 23.73 21.53 1. (CC) gcc options: -lm
Xonotic This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 3840 x 2160 - Effects Quality: Low Linux 5.10.2 Linux 5.11 Git 140 280 420 560 700 SE +/- 2.78, N = 3 SE +/- 4.89, N = 3 645.76 636.64 MIN: 390 / MAX: 1323 MIN: 392 / MAX: 1290
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 0.7064 1.4128 2.1192 2.8256 3.532 SE +/- 0.00298, N = 3 SE +/- 0.01051, N = 3 3.13966 3.13525 MIN: 3.04 MIN: 3.03 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.org ns, Fewer Is Better BenchmarkMutex Benchmark: Mutex Lock Unlock std::mutex Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 14.8 14.8 1. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Process Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 15.40 16.17 1. (CC) gcc options: -lpthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 0.8493 1.6986 2.5479 3.3972 4.2465 SE +/- 0.01821, N = 3 SE +/- 0.04653, N = 3 3.61756 3.77483 MIN: 3.36 MIN: 3.51 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads Linux 5.10.2 Linux 5.11 Git 2 4 6 8 10 SE +/- 0.064519, N = 15 SE +/- 0.106917, N = 3 8.899371 8.718967 1. (CC) gcc options: -lm
yquake2 This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: Software CPU - Resolution: 1920 x 1080 Linux 5.10.2 Linux 5.11 Git 40 80 120 160 200 SE +/- 0.85, N = 3 SE +/- 11.05, N = 15 179.1 152.9 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: Software CPU - Resolution: 3840 x 2160 Linux 5.10.2 Linux 5.11 Git 11 22 33 44 55 SE +/- 0.22, N = 3 SE +/- 0.35, N = 3 49.3 49.5 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GB/sec, More Is Better perf-bench Benchmark: Memset 1MB Linux 5.10.2 Linux 5.11 Git 6 12 18 24 30 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 26.57 26.57 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 0.2171 0.4342 0.6513 0.8684 1.0855 SE +/- 0.002969, N = 3 SE +/- 0.003270, N = 3 0.815800 0.965011 MIN: 0.79 MIN: 0.93 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
RealSR-NCNN RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better RealSR-NCNN 20200818 Scale: 4x - TAA: No Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 11.75 11.77
OpenBenchmarking.org Messages Per Second, More Is Better IPC_benchmark Type: Unnamed Pipe - Message Bytes: 4096 Linux 5.10.2 Linux 5.11 Git 500K 1000K 1500K 2000K 2500K SE +/- 16121.23, N = 15 SE +/- 24046.99, N = 3 2436716 2500263
Opus Codec Encoding Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Opus Codec Encoding 1.3.1 WAV To Opus Encode Linux 5.10.2 Linux 5.11 Git 2 4 6 8 10 SE +/- 0.068, N = 5 SE +/- 0.058, N = 5 6.106 6.134 1. (CXX) g++ options: -fvisibility=hidden -logg -lm
MPV MPV is an open-source, cross-platform media player. This test profile tests the frame-rate that can be achieved unsynchronized in a desynchronized mode. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org FPS, More Is Better MPV Video Input: Big Buck Bunny Sunflower 1080p - Decode: Software Only Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 6.87, N = 3 SE +/- 11.16, N = 3 2077.19 1955.84 MIN: 1000 / MAX: 4000.16 MIN: 923.08 / MAX: 4000.16 1. mpv 0.32.0
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 2 4 6 8 10 SE +/- 0.03242, N = 3 SE +/- 0.01612, N = 3 7.62049 8.08813 MIN: 7.32 MIN: 7.95 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Random Fill Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 83.06 80.71 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Random Fill Linux 5.10.2 Linux 5.11 Git 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 32.0 32.9 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Overwrite Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.30, N = 3 SE +/- 0.40, N = 3 83.23 80.26 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Overwrite Linux 5.10.2 Linux 5.11 Git 8 16 24 32 40 SE +/- 0.12, N = 3 SE +/- 0.19, N = 3 31.9 33.1 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org Messages Per Second, More Is Better IPC_benchmark Type: Unnamed Pipe - Message Bytes: 2048 Linux 5.10.2 Linux 5.11 Git 800K 1600K 2400K 3200K 4000K SE +/- 27847.34, N = 3 SE +/- 32697.95, N = 15 3619389 3836785
Waifu2x-NCNN Vulkan Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Waifu2x-NCNN Vulkan 20200818 Scale: 2x - Denoise: 3 - TAA: Yes Linux 5.10.2 Linux 5.11 Git 2 4 6 8 10 SE +/- 0.009, N = 3 SE +/- 0.012, N = 3 6.439 6.371
yquake2 This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: Software CPU - Resolution: 2560 x 1440 Linux 5.10.2 Linux 5.11 Git 20 40 60 80 100 SE +/- 0.60, N = 3 SE +/- 1.46, N = 3 105.3 106.3 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 15.79 16.36 MIN: 15.46 MIN: 16.01 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations Linux 5.10.2 Linux 5.11 Git 12 24 36 48 60 SE +/- 0.59, N = 4 SE +/- 0.04, N = 3 52.89 51.80 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files Linux 5.10.2 Linux 5.11 Git 3 6 9 12 15 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 10.15 10.09 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs Linux 5.10.2 Linux 5.11 Git 7 14 21 28 35 SE +/- 0.33, N = 3 SE +/- 0.23, N = 3 29.93 29.17 1. (CC) gcc options: -lm
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 2 - Type: Process Linux 5.10.2 Linux 5.11 Git 1.0323 2.0646 3.0969 4.1292 5.1615 SE +/- 0.001, N = 3 SE +/- 0.005, N = 3 4.588 4.559 1. (CC) gcc options: -lpthread
perf-bench This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ops/sec, More Is Better perf-bench Benchmark: Syscall Basic Linux 5.10.2 Linux 5.11 Git 5M 10M 15M 20M 25M SE +/- 103801.56, N = 3 SE +/- 5682.55, N = 3 22515427 22277947 1. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.0 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU Linux 5.10.2 Linux 5.11 Git 1.0805 2.161 3.2415 4.322 5.4025 SE +/- 0.01252, N = 3 SE +/- 0.01970, N = 3 4.78821 4.80215 MIN: 4.54 MIN: 4.5 1. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
LevelDB LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds Per Op, Fewer Is Better LevelDB 1.22 Benchmark: Fill Sync Linux 5.10.2 Linux 5.11 Git 400 800 1200 1600 2000 SE +/- 1.19, N = 3 SE +/- 3.72, N = 3 1781.32 1738.54 1. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.org MB/s, More Is Better LevelDB 1.22 Benchmark: Fill Sync Linux 5.10.2 Linux 5.11 Git 0.3375 0.675 1.0125 1.35 1.6875 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.5 1.5 1. (CXX) g++ options: -O3 -lsnappy -lpthread
yquake2 This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: OpenGL 3.x - Resolution: 2560 x 1440 Linux 5.10.2 Linux 5.11 Git 200 400 600 800 1000 SE +/- 1.50, N = 3 SE +/- 7.43, N = 3 976.8 975.4 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: OpenGL 3.x - Resolution: 1920 x 1080 Linux 5.10.2 Linux 5.11 Git 200 400 600 800 1000 SE +/- 1.84, N = 3 SE +/- 0.00, N = 3 980.4 982.9 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.org Frames Per Second, More Is Better yquake2 7.45 Renderer: OpenGL 3.x - Resolution: 3840 x 2160 Linux 5.10.2 Linux 5.11 Git 200 400 600 800 1000 SE +/- 1.84, N = 3 SE +/- 6.17, N = 3 980.8 980.4 1. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
HPC Challenge HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Max Ping Pong Bandwidth Linux 5.10.2 Linux 5.11 Git 7K 14K 21K 28K 35K SE +/- 107.50, N = 3 SE +/- 161.16, N = 3 32633.79 29671.38 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Bandwidth Linux 5.10.2 Linux 5.11 Git 0.9518 1.9036 2.8554 3.8072 4.759 SE +/- 0.01900, N = 3 SE +/- 0.07138, N = 3 4.23009 4.12610 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org usecs, Fewer Is Better HPC Challenge 1.5.0 Test / Class: Random Ring Latency Linux 5.10.2 Linux 5.11 Git 0.0988 0.1976 0.2964 0.3952 0.494 SE +/- 0.00200, N = 3 SE +/- 0.00254, N = 3 0.43563 0.43915 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org GUP/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Random Access Linux 5.10.2 Linux 5.11 Git 0.0104 0.0208 0.0312 0.0416 0.052 SE +/- 0.00030, N = 3 SE +/- 0.00016, N = 3 0.04604 0.04635 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: EP-STREAM Triad Linux 5.10.2 Linux 5.11 Git 0.4916 0.9832 1.4748 1.9664 2.458 SE +/- 0.00104, N = 3 SE +/- 0.00009, N = 3 2.17512 2.18509 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org GB/s, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ptrans Linux 5.10.2 Linux 5.11 Git 0.6673 1.3346 2.0019 2.6692 3.3365 SE +/- 0.00244, N = 3 SE +/- 0.00150, N = 3 2.96569 2.96599 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: EP-DGEMM Linux 5.10.2 Linux 5.11 Git 5 10 15 20 25 SE +/- 0.13, N = 3 SE +/- 0.10, N = 3 19.20 19.40 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.org GFLOPS, More Is Better HPC Challenge 1.5.0 Test / Class: G-Ffte Linux 5.10.2 Linux 5.11 Git 2 4 6 8 10 SE +/- 0.07039, N = 3 SE +/- 0.06723, N = 3 6.02335 6.03540 1. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops 2. OpenBLAS + Open MPI 4.0.3
Linux 5.10.2 Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (2702 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211
OS: Ubuntu 20.10, Kernel: 5.10.2-051002-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009Graphics Notes: GLAMORPython Notes: Python 3.8.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 22 December 2020 10:32 by user phoronix.
Linux 5.11 Git Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (2702 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211
OS: Ubuntu 20.10, Kernel: 5.10.0-051000daily20201222-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009Graphics Notes: GLAMORPython Notes: Python 3.8.6Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 23 December 2020 05:56 by user phoronix.