Graviton4 vs. Graviton3

Benchmarks by Michael Larabel

HTML result view exported from: https://openbenchmarking.org/result/2407104-NE-2407106NE45&grs&rdt.

Graviton4 vs. Graviton3ProcessorMotherboardChipsetMemoryDiskNetworkOSKernelCompilerFile-SystemSystem LayerGraviton4 r8g.16xlargeGraviton3 r7g.16xlargeARMv8 Neoverse-V2 (64 Cores)Amazon EC2 r8g.16xlarge (1.0 BIOS)Amazon Device 0200496GB429GB Amazon Elastic Block StoreAmazon ElasticUbuntu 24.046.8.0-1009-aws (aarch64)GCC 13.2.0ext4amazonARMv8 Neoverse-V1 (64 Cores)Amazon EC2 r7g.16xlarge (1.0 BIOS)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v Python Details- Python 3.12.3Security Details- Graviton4 r8g.16xlarge: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected - Graviton3 r7g.16xlarge: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of CSV2 BHB + srbds: Not affected + tsx_async_abort: Not affected

Graviton4 vs. Graviton3xmrig: GhostRider - 1Mminife: Smallpgbench: 100 - 1000 - Read Onlypgbench: 100 - 1000 - Read Only - Average Latencyincompact3d: input.i3d 193 Cells Per Directionopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Execution Timebuild-nodejs: Time To Compilerocksdb: Update Randxmrig: KawPow - 1Mxmrig: CryptoNight-Heavy - 1Mxmrig: CryptoNight-Femto UPX2 - 1Mxmrig: Monero - 1Mrocksdb: Rand Readbuild-llvm: Ninjaxmrig: Wownero - 1Mc-ray: 5K - 16c-ray: 4K - 16openfoam: drivaerFastback, Small Mesh Size - Mesh Timebuild-gem5: Time To Compilebuild-godot: Time To Compileopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timerocksdb: Read Rand Write Randclickhouse: 100M Rows Hits Dataset, Third Runcompress-7zip: Compression Ratingblender: Barbershop - CPU-Onlyclickhouse: 100M Rows Hits Dataset, First Run / Cold Cacheclickhouse: 100M Rows Hits Dataset, Second Runliquid-dsp: 64 - 256 - 32srsran: PUSCH Processor Benchmark, Throughput Totaljohn-the-ripper: HMAC-SHA512rocksdb: Read While Writingblender: BMW27 - CPU-Onlyblender: Classroom - CPU-Onlyblender: Pabellon Barcelona - CPU-Onlyblender: Fishy Cat - CPU-Onlyopenssl: SHA256liquid-dsp: 64 - 256 - 57compress-7zip: Decompression Ratinggromacs: MPI CPU - water_GMX50_bareopenssl: SHA512srsran: PDSCH Processor Benchmark, Throughput Totaljohn-the-ripper: Blowfishjohn-the-ripper: bcryptliquid-dsp: 64 - 256 - 512john-the-ripper: WPA PSKjohn-the-ripper: MD5openssl: ChaCha20-Poly1305openssl: ChaCha20openssl: AES-256-GCMopenssl: AES-128-GCMpgbench: 100 - 1000 - Read Writepgbench: 100 - 1000 - Read Write - Average Latencystockfish: Chess BenchmarkGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge5958.765410.419475250.5147.8750206641.413075344.1676365.024127461621906.021867.221851.921872.1346532586182.06328304.050.24728.32318.384768186.768147.87091.9745275584014495.03383460499.63449.44479.9432587333331332.5108066333852259750.64105.39202.2695.015697117453719296666673315054.8313541280936314402.55703257038185000000574441579667751905487071018987203232661251505272997971021374420226.244814408012108.536361.710874880.92013.955848468.294615552.5657571.01181747014540.414553.914547.514564.0243166771257.85320409.668.69938.66824.784465251.402195.497121.278614254497382.14296731644.98348.41375.2925738000001053.886689000687238962.77129.48247.60115.474739551236316547333332845764.1633065946571012612.6506795070916631333352631145133369430704540941475020772490955005972875915247634541220.27355548662OpenBenchmarking.org

Xmrig

Variant: GhostRider - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge13002600390052006500SE +/- 47.30, N = 12SE +/- 4.85, N = 35958.72108.51. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

miniFE

Problem Size: Small

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge14K28K42K56K70KSE +/- 28.12, N = 3SE +/- 21.14, N = 365410.436361.71. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Only

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read OnlyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge400K800K1200K1600K2000KSE +/- 6233.51, N = 3SE +/- 5453.81, N = 3194752510874881. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average LatencyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge0.2070.4140.6210.8281.035SE +/- 0.002, N = 3SE +/- 0.005, N = 30.5140.9201. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per DirectionGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge48121620SE +/- 0.01884341, N = 3SE +/- 0.01649632, N = 37.8750206613.955848401. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Execution Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution TimeGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge153045607541.4168.291. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Execution Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution TimeGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge120240360480600344.17552.571. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 21.7.2Time To CompileGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge120240360480600SE +/- 0.19, N = 3SE +/- 0.62, N = 3365.02571.01

RocksDB

Test: Update Random

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Update RandomGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge300K600K900K1200K1500KSE +/- 11457.12, N = 3SE +/- 7555.40, N = 1512746168174701. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Xmrig

Variant: KawPow - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: KawPow - Hash Count: 1MGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge5K10K15K20K25KSE +/- 59.25, N = 3SE +/- 6.29, N = 321906.014540.41. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Xmrig

Variant: CryptoNight-Heavy - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: CryptoNight-Heavy - Hash Count: 1MGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge5K10K15K20K25KSE +/- 16.81, N = 3SE +/- 14.63, N = 321867.214553.91. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Xmrig

Variant: CryptoNight-Femto UPX2 - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: CryptoNight-Femto UPX2 - Hash Count: 1MGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge5K10K15K20K25KSE +/- 9.16, N = 3SE +/- 4.95, N = 321851.914547.51. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

Xmrig

Variant: Monero - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: Monero - Hash Count: 1MGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge5K10K15K20K25KSE +/- 40.15, N = 3SE +/- 12.04, N = 321872.114564.01. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

RocksDB

Test: Random Read

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Random ReadGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge70M140M210M280M350MSE +/- 1371022.17, N = 3SE +/- 36358.69, N = 33465325862431667711. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Timed LLVM Compilation

Build System: Ninja

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge60120180240300SE +/- 0.20, N = 3SE +/- 0.38, N = 3182.06257.85

Xmrig

Variant: Wownero - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: Wownero - Hash Count: 1MGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge6K12K18K24K30KSE +/- 16.31, N = 3SE +/- 8.34, N = 328304.020409.61. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

C-Ray

Resolution: 5K - Rays Per Pixel: 16

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 5K - Rays Per Pixel: 16Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge1530456075SE +/- 0.07, N = 3SE +/- 0.01, N = 350.2568.701. (CC) gcc options: -lpthread -lm

C-Ray

Resolution: 4K - Rays Per Pixel: 16

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 2.0Resolution: 4K - Rays Per Pixel: 16Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge918273645SE +/- 0.03, N = 3SE +/- 0.04, N = 328.3238.671. (CC) gcc options: -lpthread -lm

OpenFOAM

Input: drivaerFastback, Small Mesh Size - Mesh Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge61218243018.3824.781. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

Timed Gem5 Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 23.0.1Time To CompileGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge50100150200250SE +/- 1.43, N = 12SE +/- 2.33, N = 3186.77251.40

Timed Godot Game Engine Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge4080120160200SE +/- 0.35, N = 3SE +/- 0.85, N = 3147.87195.50

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Mesh Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge30609012015091.97121.281. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm

RocksDB

Test: Read Random Write Random

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read Random Write RandomGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge1.2M2.4M3.6M4.8M6MSE +/- 11170.68, N = 3SE +/- 1357.98, N = 3558401442544971. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

ClickHouse

100M Rows Hits Dataset, Third Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge110220330440550SE +/- 6.71, N = 9SE +/- 2.76, N = 9495.03382.14MIN: 43.07 / MAX: 6666.67MIN: 33.22 / MAX: 4615.38

7-Zip Compression

Test: Compression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 24.05Test: Compression RatingGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge80K160K240K320K400KSE +/- 397.84, N = 3SE +/- 546.41, N = 33834602967311. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Blender

Blend File: Barbershop - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: Barbershop - Compute: CPU-OnlyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge140280420560700SE +/- 0.76, N = 3SE +/- 0.24, N = 3499.63644.98

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge100200300400500SE +/- 4.96, N = 9SE +/- 5.68, N = 9449.44348.41MIN: 42.7 / MAX: 6666.67MIN: 33.19 / MAX: 5000

ClickHouse

100M Rows Hits Dataset, Second Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge100200300400500SE +/- 5.05, N = 9SE +/- 3.03, N = 9479.94375.29MIN: 43.1 / MAX: 6000MIN: 33.15 / MAX: 5000

Liquid-DSP

Threads: 64 - Buffer Length: 256 - Filter Length: 32

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 32Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge700M1400M2100M2800M3500MSE +/- 533333.33, N = 3SE +/- 404145.19, N = 3325873333325738000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Total

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput TotalGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge30060090012001500SE +/- 0.03, N = 3SE +/- 0.03, N = 31332.51053.8MIN: 784 / MAX: 1332.6MIN: 599.4 / MAX: 1053.91. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

John The Ripper

Test: HMAC-SHA512

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: HMAC-SHA512Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge20M40M60M80M100MSE +/- 640028.21, N = 3SE +/- 76622.45, N = 3108066333866890001. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

RocksDB

Test: Read While Writing

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While WritingGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge2M4M6M8M10MSE +/- 26974.32, N = 3SE +/- 6592.32, N = 3852259768723891. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Blender

Blend File: BMW27 - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: BMW27 - Compute: CPU-OnlyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge1428425670SE +/- 0.10, N = 3SE +/- 0.02, N = 350.6462.77

Blender

Blend File: Classroom - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: Classroom - Compute: CPU-OnlyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge306090120150SE +/- 0.11, N = 3SE +/- 0.06, N = 3105.39129.48

Blender

Blend File: Pabellon Barcelona - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: Pabellon Barcelona - Compute: CPU-OnlyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge50100150200250SE +/- 0.34, N = 3SE +/- 0.13, N = 3202.26247.60

Blender

Blend File: Fishy Cat - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.0.2Blend File: Fishy Cat - Compute: CPU-OnlyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge306090120150SE +/- 0.26, N = 3SE +/- 0.23, N = 395.01115.47

OpenSSL

Algorithm: SHA256

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: SHA256Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge12000M24000M36000M48000M60000MSE +/- 10002237.60, N = 3SE +/- 29159419.58, N = 356971174537473955123631. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

Liquid-DSP

Threads: 64 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 57Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge400M800M1200M1600M2000MSE +/- 33333.33, N = 3SE +/- 133333.33, N = 3192966666716547333331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

7-Zip Compression

Test: Decompression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 24.05Test: Decompression RatingGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge70K140K210K280K350KSE +/- 107.45, N = 3SE +/- 70.02, N = 33315052845761. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bareGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge1.0872.1743.2614.3485.435SE +/- 0.004, N = 3SE +/- 0.001, N = 34.8314.1631. (CXX) g++ options: -O3 -lm

OpenSSL

Algorithm: SHA512

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: SHA512Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge8000M16000M24000M32000M40000MSE +/- 8695334.33, N = 3SE +/- 3137583.25, N = 335412809363306594657101. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

srsRAN Project

Test: PDSCH Processor Benchmark, Throughput Total

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput TotalGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge3K6K9K12K15KSE +/- 34.15, N = 3SE +/- 20.13, N = 314402.512612.61. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge12K24K36K48K60KSE +/- 9.29, N = 3SE +/- 15.39, N = 357032506791. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

John The Ripper

Test: bcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge12K24K36K48K60KSE +/- 2.33, N = 3SE +/- 8.67, N = 357038507091. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

Liquid-DSP

Threads: 64 - Buffer Length: 256 - Filter Length: 512

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 512Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge40M80M120M160M200MSE +/- 0.00, N = 3SE +/- 3333.33, N = 31850000001663133331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

John The Ripper

Test: WPA PSK

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge12K24K36K48K60KSE +/- 10.97, N = 3SE +/- 0.00, N = 357444526311. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

John The Ripper

Test: MD5

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge300K600K900K1200K1500KSE +/- 1201.85, N = 3SE +/- 333.33, N = 3157966714513331. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenSSL

Algorithm: ChaCha20-Poly1305

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: ChaCha20-Poly1305Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge16000M32000M48000M64000M80000MSE +/- 558939.35, N = 3SE +/- 6321669.59, N = 375190548707694307045401. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenSSL

Algorithm: ChaCha20

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: ChaCha20Graviton4 r8g.16xlargeGraviton3 r7g.16xlarge20000M40000M60000M80000M100000MSE +/- 66408.09, N = 3SE +/- 4627810.23, N = 3101898720323941475020771. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenSSL

Algorithm: AES-256-GCM

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: AES-256-GCMGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge60000M120000M180000M240000M300000MSE +/- 6417811.59, N = 3SE +/- 4988409.71, N = 32661251505272490955005971. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

OpenSSL

Algorithm: AES-128-GCM

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSLAlgorithm: AES-128-GCMGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge60000M120000M180000M240000M300000MSE +/- 7802917.99, N = 3SE +/- 3618954.21, N = 32997971021372875915247631. OpenSSL 3.0.13 30 Jan 2024 (Library: OpenSSL 3.0.13 30 Jan 2024)

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read WriteGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge10002000300040005000SE +/- 14.53, N = 3SE +/- 45.23, N = 3442045411. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

PostgreSQL

Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average Latency

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Write - Average LatencyGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge50100150200250SE +/- 0.74, N = 3SE +/- 2.18, N = 3226.24220.271. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Stockfish

Chess Benchmark

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess BenchmarkGraviton4 r8g.16xlargeGraviton3 r7g.16xlarge20M40M60M80M100MSE +/- 2235406.92, N = 15SE +/- 1152869.30, N = 1281440801555486621. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -flto -flto-partition=one -flto=jobserver

Phoronix Test Suite v10.8.5