Amazon EC2 c7g.4xlarge Graviton3

Graviton3 benchmarks by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/2206141-NE-2206134NE90&grs.

Amazon EC2 c7g.4xlarge Graviton3ProcessorMotherboardChipsetMemoryDiskNetworkGraphicsOSKernelCompilerFile-SystemSystem LayerDisplay ServerVulkanc7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30ARMv8 Neoverse-V1 (16 Cores)Amazon EC2 c7g.4xlarge (1.0 BIOS)Amazon Device 020032GB193GB Amazon Elastic Block StoreAmazon ElasticUbuntu 22.045.15.0-1004-aws (aarch64)GCC 11.2.0ext4amazonARMv8 Neoverse-N1 (16 Cores)QEMU KVM Virtual Machine (0.0.0 BIOS)Red Hat QEMU PCIe1 x 7000 MB RAM QEMU47GBRed Hat Virtio deviceUbuntu 20.045.4.0-100-generic (aarch64)X Server 1.20.131.1.182GCC 9.4.0KVM16384 MB + 15031 MB RAM5.4.0-117-generic (aarch64)8 x Intel Core (Haswell no TSX) (8 Cores)OpenStack Foundation Nova v19.0.4 (2:1.10.2-58953eb7 BIOS)Intel 440FX 82441FX PMC16384 MB + 13616 MB RAM215GB QEMU HDDCirrus Logic GD 54462 x Red Hat Virtio device5.4.0-99-generic (x86_64) 20220203OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- c7g.4xlarge: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-bootstrap --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-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto -v - 16 vcpu ampere Vm: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --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 -v - 16 vcpu ampere Vm run 2: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --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 -v - pinned 16 vcpu ampere Vm: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --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 -v - b2-30: --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-Av3uEd/gcc-9-9.4.0/debian/tmp-nvptx/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 -v Java Details- c7g.4xlarge: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)- 16 vcpu ampere Vm: OpenJDK Runtime Environment (build 11.0.13+8-Ubuntu-0ubuntu1.20.04)- 16 vcpu ampere Vm run 2: OpenJDK Runtime Environment (build 11.0.13+8-Ubuntu-0ubuntu1.20.04)- pinned 16 vcpu ampere Vm: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.20.04.1)- b2-30: OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.20.04.1)Python Details- c7g.4xlarge: Python 3.10.4- 16 vcpu ampere Vm: Python 3.8.10- 16 vcpu ampere Vm run 2: Python 3.8.10- pinned 16 vcpu ampere Vm: Python 3.8.10- b2-30: Python 3.8.10Security Details- c7g.4xlarge: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: 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- 16 vcpu ampere Vm: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: 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- 16 vcpu ampere Vm run 2: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: 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- pinned 16 vcpu ampere Vm: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: 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 - b2-30: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT disabled + mds: Vulnerable: Clear buffers attempted no microcode; SMT Host state unknown + meltdown: Mitigation of PTI + spec_store_bypass: Vulnerable + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline STIBP: disabled RSB filling + srbds: Unknown: Dependent on hypervisor status + tsx_async_abort: Not affected Processor Details- b2-30: CPU Microcode: 0x1

Amazon EC2 c7g.4xlarge Graviton3openssl: SHA256nginx: 1000nginx: 200nginx: 100nginx: 500mt-dgemm: Sustained Floating-Point Ratetensorflow-lite: Mobilenet Quantnpb: IS.Ddacapobench: Tradesoapcompress-zstd: 19, Long Mode - Compression Speedtensorflow-lite: NASNet Mobilerodinia: OpenMP CFD Solverincompact3d: input.i3d 193 Cells Per Directionamg: incompact3d: input.i3d 129 Cells Per Directionrodinia: OpenMP LavaMDc-ray: Total Time - 4K, 16 Rays Per Pixelapache: 100npb: LU.Clulesh: apache: 1000apache: 200apache: 500compress-7zip: Decompression Ratingdacapobench: H2openssl: RSA4096lammps: 20k Atomslammps: Rhodopsin Proteinbuild-llvm: Ninjaopenssl: RSA4096build-nodejs: Time To Compileasmfish: 1024 Hash Memory, 26 Depthbuild-gem5: Time To Compilecoremark: CoreMark Size 666 - Iterations Per Secondn-queens: Elapsed Timeavifenc: 6povray: Trace Timem-queens: Time To Solveavifenc: 6, Losslesstensorflow-lite: Inception ResNet V2stockfish: Total Timenpb: MG.Cwebp: Quality 100, Lossless, Highest Compressionwebp: Quality 100, Losslesstensorflow-lite: Inception V4tensorflow-lite: Mobilenet Floatnpb: SP.Clczero: Eigenavifenc: 10, Losslessbuild2: Time To Compiletensorflow-lite: SqueezeNetsimdjson: DistinctUserIDstress-ng: Memory Copyingbuild-apache: Time To Compilelczero: BLASgpaw: Carbon Nanotubedacapobench: Jythonsimdjson: PartialTweetsgromacs: MPI CPU - water_GMX50_barecompress-zstd: 19, Long Mode - Decompression Speedcompress-zstd: 3 - Decompression Speedmrbayes: Primate Phylogeny Analysisquantlib: avifenc: 2simdjson: Kostyatscp: AI Chess Performancenpb: BT.Cphpbench: PHP Benchmark Suiteliquid-dsp: 16 - 256 - 57npb: EP.Dastcenc: Thoroughstress-ng: Vector Mathsynthmark: VoiceMark_100astcenc: Exhaustivesecuremark: SecureMark-TLSsimdjson: LargeRandstress-ng: CPU Stresswebp: Quality 100, Highest Compressionpybench: Total For Average Test Timesstress-ng: Cryptostress-ng: Matrix Mathonnx: super-resolution-10 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: GPT-2 - CPU - Standardstress-ng: IO_uringstress-ng: CPU Cachengspice: C7552ngspice: C2670build-php: Time To Compilebuild-imagemagick: Time To Compileavifenc: 0compress-7zip: Compression Ratingcompress-zstd: 19 - Decompression Speedcompress-zstd: 19 - Compression Speedcompress-zstd: 3 - Compression Speeddacapobench: Tradebeansrodinia: OpenMP Streamclusternpb: FT.Cnpb: CG.Chpcg: c7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3013722045973346814.75352380.98345710.87346613.345.8538641502.951041.90352439.511591.910.47829.125857012588073338.01671425143.33438.51767231.887730.4110940.93972719.3373676.9573546.327305429512546.411.42511.291544.929178460.4497.57932134123391.171405413.86055421.5369.38537.86366.82211.90840051.32760889113481.6148.20822.76941855.12156.604467.1911895.765115.0203257.942.696693.3226.9401103155.18039402.621.1283240.63508.5251.3972512.7141.6981.94137009410339.53666484383606667934.7213.924855258.17675.635139.37971837080.75029.719.346118523181.8180088.742817609384077990843015.7864.31191.286198.22469.48327.904256.841978243050.341.24639.1320313.29611791.776571.9526.30581878.46924.50661.87650.952651.141275370693064815.2466077.4865315.0665751.851.1479445538.34227.19161268.8950384.927.85258.256130256647790017.9038974190.40054.10217823.745892.123014.474520049.5220474.1321329.88581669907791.05.5865.4951429.67264694.31386.10516217696960.521371146.66303312.55112.55569.75864.40015.04898426.5159709776924.28125.47058.69187437.75496.112009.154857.877235.4967508.831.663189.3155.985555305.71767531.650.6101904.52095.8339.7231910.9223.6011.3010403886799.06491649319180000660.3819.071841006.21559.893183.50301418780.553974.4810.299142120420.0976853.01236.50796.446654.186162.60876.133419.324406061449.114.01063.32443652.3886338.872436.269.2600312781197210311688.57340989.94343517.35327213.694.1819121802.94385.47540032.014522.113.68440.802258889986706711.1243341188.94251.07459296.376106.086092.607160484.6066344.0565472.33705004283791.79.0068.806724.16364764.0651.03329098784520.876373066.69931612.52311.46551.16464.30913.84243137.5240785797988.6957.45727.26544528.72283.212767.508777.184126.1393665.991.873740.2831.353946202.17848851.830.8582466.12825.7323.1171975.5209.2121.4710407787720.30491470319556667671.0816.892641030.72561.231156.81111417330.574016.6110.098142520364.2376899.28213.47239.738239.23982.03538.645356.245806142343.537.82938.6472813.3497344.743826.4817.644916563885701.8882717113.04343.69864015.721757.445.041125.16722630655738032.1106580572.565150.54721501.084058.9773214555586918.93.5783.6421671.37360337.11447.825110879401104.322146387.22251634.63625.458102.703172.31331.6691063511043658318048.3259.33927.5751070535135.985043.8595113.709270.0097230.653.6043.90198375933.110.6041762.71954.9198.1461580.5183.3662.058804769055.91445360266883333713.8215.6691504.353140.19521472680.6711.3401413198.368254.644122.81063.356396.307284371736.819.01405.5757230.8889878.382958.095.93594OpenBenchmarking.org

OpenSSL

Algorithm: SHA256

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.0Algorithm: SHA256c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-303000M6000M9000M12000M15000MSE +/- 7739237.92, N = 3SE +/- 12960363.63, N = 3SE +/- 12263977.94, N = 3SE +/- 4321270.11, N = 313722045973127537069301278119721016563885701. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

nginx

Concurrent Requests: 1000

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1000c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm70K140K210K280K350KSE +/- 1410.11, N = 3SE +/- 705.50, N = 3SE +/- 1167.75, N = 3346814.7564815.24311688.57-ldl -lpthread-ldl -lpthread1. (CC) gcc options: -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 200

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 200c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm80K160K240K320K400KSE +/- 3986.77, N = 3SE +/- 315.71, N = 3SE +/- 353.86, N = 3352380.9866077.48340989.94-ldl -lpthread1. (CC) gcc options: -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 100

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 100c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm70K140K210K280K350KSE +/- 2009.97, N = 3SE +/- 90.74, N = 3SE +/- 1154.62, N = 3345710.8765315.06343517.35-ldl -lpthread-ldl -lpthread1. (CC) gcc options: -lcrypt -lz -O3 -march=native

nginx

Concurrent Requests: 500

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 500c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm70K140K210K280K350KSE +/- 1017.52, N = 3SE +/- 465.70, N = 3SE +/- 556.31, N = 3346613.3465751.85327213.69-ldl -lpthread-ldl -lpthread1. (CC) gcc options: -lcrypt -lz -O3 -march=native

ACES DGEMM

Sustained Floating-Point Rate

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point Ratec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-301.31712.63423.95135.26846.5855SE +/- 0.016350, N = 3SE +/- 0.004261, N = 3SE +/- 0.044251, N = 5SE +/- 0.017946, N = 35.8538641.1479444.1819121.8882711. (CC) gcc options: -O3 -march=native -fopenmp

TensorFlow Lite

Model: Mobilenet Quant

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet Quantc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3015003000450060007500SE +/- 17.76, N = 3SE +/- 44.20, N = 15SE +/- 9.55, N = 3SE +/- 41.00, N = 31502.955538.341802.947113.04

NAS Parallel Benchmarks

Test / Class: IS.D

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.Dc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302004006008001000SE +/- 2.29, N = 3SE +/- 2.75, N = 4SE +/- 0.61, N = 3SE +/- 2.93, N = 81041.90227.19385.47343.69-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

DaCapo Benchmark

Java Test: Tradesoap

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Tradesoapc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-303K6K9K12K15KSE +/- 14.95, N = 4SE +/- 114.92, N = 4SE +/- 80.76, N = 16SE +/- 93.14, N = 435241612654008640

Zstd Compression

Compression Level: 19, Long Mode - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression Speedc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30918273645SE +/- 0.23, N = 3SE +/- 0.11, N = 3SE +/- 0.03, N = 3SE +/- 0.20, N = 339.508.8932.0015.70-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

TensorFlow Lite

Model: NASNet Mobile

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: NASNet Mobilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3011K22K33K44K55KSE +/- 121.56, N = 15SE +/- 645.70, N = 15SE +/- 128.17, N = 7SE +/- 185.64, N = 1511591.950384.914522.121757.4

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD Solverc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-301020304050SE +/- 0.02, N = 3SE +/- 0.28, N = 15SE +/- 0.03, N = 3SE +/- 0.42, N = 310.4827.8513.6845.041. (CXX) g++ options: -O2 -lOpenCL

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per Directionc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30306090120150SE +/- 0.03, N = 3SE +/- 0.42, N = 3SE +/- 0.22, N = 3SE +/- 1.07, N = 329.1358.2640.80125.17-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi

Algebraic Multi-Grid Benchmark

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30300M600M900M1200M1500MSE +/- 952437.28, N = 3SE +/- 1444853.58, N = 3SE +/- 10562657.64, N = 3SE +/- 3392944.23, N = 51258807333566477900899867067306557380-pthread-pthread-pthread1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

Xcompact3d Incompact3d

Input: input.i3d 129 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per Directionc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30714212835SE +/- 0.01401446, N = 3SE +/- 0.21523497, N = 15SE +/- 0.10843597, N = 14SE +/- 0.09399703, N = 38.0167142517.9038974011.1243341032.11065800-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -cpp -O2 -funroll-loops -floop-optimize -fcray-pointer -fbacktrace -lmpi_usempif08 -lmpi_mpifh -lmpi

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30120240360480600SE +/- 0.15, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 1.27, N = 3143.33190.40188.94572.571. (CXX) g++ options: -O2 -lOpenCL

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per Pixelc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30306090120150SE +/- 0.02, N = 3SE +/- 0.42, N = 3SE +/- 0.03, N = 3SE +/- 0.36, N = 338.5254.1051.07150.551. (CC) gcc options: -lm -lpthread -O3

Apache HTTP Server

Concurrent Requests: 100

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 100c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm14K28K42K56K70KSE +/- 38.09, N = 3SE +/- 99.44, N = 3SE +/- 154.57, N = 367231.8817823.7459296.37-pthread-pthread1. (CC) gcc options: -shared -fPIC -O2

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.Cc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-305K10K15K20K25KSE +/- 1.96, N = 3SE +/- 16.78, N = 3SE +/- 14.27, N = 3SE +/- 29.19, N = 37730.415892.126106.0821501.08-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

LULESH

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302K4K6K8K10KSE +/- 76.73, N = 3SE +/- 45.42, N = 15SE +/- 57.18, N = 7SE +/- 56.60, N = 310940.943014.476092.614058.98-pthread-pthread-pthread1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

Apache HTTP Server

Concurrent Requests: 1000

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 1000c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm16K32K48K64K80KSE +/- 83.83, N = 3SE +/- 134.65, N = 3SE +/- 232.71, N = 372719.3320049.5260484.60-pthread-pthread1. (CC) gcc options: -shared -fPIC -O2

Apache HTTP Server

Concurrent Requests: 200

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 200c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm16K32K48K64K80KSE +/- 649.31, N = 3SE +/- 231.23, N = 3SE +/- 237.38, N = 373676.9520474.1366344.05-pthread-pthread1. (CC) gcc options: -shared -fPIC -O2

Apache HTTP Server

Concurrent Requests: 500

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 500c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm16K32K48K64K80KSE +/- 89.82, N = 3SE +/- 197.16, N = 3SE +/- 373.57, N = 373546.3221329.8865472.33-pthread-pthread1. (CC) gcc options: -shared -fPIC -O2

7-Zip Compression

Test: Decompression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 21.06Test: Decompression Ratingc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3016K32K48K64K80KSE +/- 12.88, N = 3SE +/- 100.59, N = 3SE +/- 144.46, N = 3730545816670500214551. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

DaCapo Benchmark

Java Test: H2

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302K4K6K8K10KSE +/- 32.57, N = 5SE +/- 58.80, N = 20SE +/- 44.47, N = 20SE +/- 51.21, N = 202951990742835586

OpenSSL

Algorithm: RSA4096

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-305001000150020002500SE +/- 0.23, N = 3SE +/- 0.30, N = 3SE +/- 0.42, N = 3SE +/- 7.31, N = 32546.4791.0791.7918.9-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k Atomsc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-303691215SE +/- 0.067, N = 3SE +/- 0.038, N = 3SE +/- 0.003, N = 311.4255.5869.0063.578-pthread-pthread-pthread1. (CXX) g++ options: -O3 -lm

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin Proteinc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-303691215SE +/- 0.060, N = 3SE +/- 0.039, N = 15SE +/- 0.008, N = 3SE +/- 0.022, N = 311.2915.4958.8063.642-pthread-pthread-pthread1. (CXX) g++ options: -O3 -lm

Timed LLVM Compilation

Build System: Ninja

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 13.0Build System: Ninjac7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30400800120016002000SE +/- 5.19, N = 3SE +/- 14.14, N = 3SE +/- 0.75, N = 3SE +/- 0.66, N = 3544.931429.67724.161671.37

OpenSSL

Algorithm: RSA4096

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3040K80K120K160K200KSE +/- 82.61, N = 3SE +/- 11.12, N = 3SE +/- 0.92, N = 3SE +/- 172.80, N = 3178460.464694.364764.060337.1-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 17.3Time To Compilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3030060090012001500SE +/- 2.06, N = 3SE +/- 7.93, N = 3SE +/- 1.61, N = 3SE +/- 1.83, N = 3497.581386.11651.031447.83

asmFish

1024 Hash Memory, 26 Depth

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Depthc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-307M14M21M28M35MSE +/- 104795.40, N = 3SE +/- 76504.07, N = 3SE +/- 414842.37, N = 3SE +/- 107879.50, N = 632134123162176962909878411087940

Timed Gem5 Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To Compilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302004006008001000SE +/- 1.33, N = 3SE +/- 11.13, N = 4SE +/- 0.59, N = 3SE +/- 12.03, N = 4391.17960.52520.881104.32

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Secondc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3090K180K270K360K450KSE +/- 3211.91, N = 3SE +/- 755.08, N = 3SE +/- 85.89, N = 3SE +/- 547.38, N = 3405413.86371146.66373066.70146387.221. (CC) gcc options: -O2 -lrt" -lrt

N-Queens

Elapsed Time

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed Timec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30816243240SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 321.5412.5512.5234.641. (CC) gcc options: -static -fopenmp -O3 -march=native

libavif avifenc

Encoder Speed: 6

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30612182430SE +/- 0.025, N = 3SE +/- 0.026, N = 3SE +/- 0.032, N = 3SE +/- 0.183, N = 39.38512.55511.46525.4581. (CXX) g++ options: -O3 -fPIC -lm

POV-Ray

Trace Time

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace Timec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3020406080100SE +/- 0.01, N = 3SE +/- 0.73, N = 15SE +/- 0.56, N = 4SE +/- 0.44, N = 337.8669.7651.16102.70-R/usr/lib-pthread-pthread-march=native -pthread1. (CXX) g++ options: -pipe -O3 -ffast-math -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

m-queens

Time To Solve

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.2Time To Solvec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-304080120160200SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.25, N = 366.8264.4064.31172.311. (CXX) g++ options: -fopenmp -O2 -march=native

libavif avifenc

Encoder Speed: 6, Lossless

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6, Losslessc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30714212835SE +/- 0.01, N = 3SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.09, N = 311.9115.0513.8431.671. (CXX) g++ options: -O3 -fPIC -lm

TensorFlow Lite

Model: Inception ResNet V2

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception ResNet V2c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3020K40K60K80K100KSE +/- 305.31, N = 3SE +/- 824.24, N = 15SE +/- 106.73, N = 3SE +/- 179.72, N = 340051.398426.543137.5106351.0

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total Timec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-306M12M18M24M30MSE +/- 153578.64, N = 3SE +/- 159724.07, N = 3SE +/- 240432.70, N = 6SE +/- 97424.64, N = 327608891159709772407857910436583-m64 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi21. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fprofile-use -fno-peel-loops -fno-tracer -pedantic -O3 -flto -flto=jobserver

NAS Parallel Benchmarks

Test / Class: MG.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.Cc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-304K8K12K16K20KSE +/- 4.69, N = 3SE +/- 12.27, N = 3SE +/- 3.72, N = 3SE +/- 64.83, N = 313481.616924.287988.6918048.32-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

WebP Image Encode

Encode Settings: Quality 100, Lossless, Highest Compression

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest Compressionc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30306090120150SE +/- 0.01, N = 3SE +/- 0.91, N = 11SE +/- 0.06, N = 3SE +/- 0.68, N = 348.21125.4757.4659.34-pthread -ltiff-pthread -ltiff-pthread -ltiff1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16

WebP Image Encode

Encode Settings: Quality 100, Lossless

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Losslessc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-301326395265SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.30, N = 422.7758.6927.2727.58-pthread -ltiff-pthread -ltiff-pthread -ltiff1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16

TensorFlow Lite

Model: Inception V4

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception V4c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3020K40K60K80K100KSE +/- 210.27, N = 3SE +/- 1013.94, N = 4SE +/- 16.95, N = 3SE +/- 345.20, N = 341855.187437.744528.7107053.0

TensorFlow Lite

Model: Mobilenet Float

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet Floatc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3012002400360048006000SE +/- 19.61, N = 3SE +/- 51.89, N = 6SE +/- 4.42, N = 3SE +/- 6.09, N = 32156.605496.112283.215135.98

NAS Parallel Benchmarks

Test / Class: SP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.Cc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3011002200330044005500SE +/- 9.61, N = 3SE +/- 8.85, N = 3SE +/- 1.24, N = 3SE +/- 17.88, N = 34467.192009.152767.505043.85-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

LeelaChessZero

Backend: Eigen

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: Eigenc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3030060090012001500SE +/- 9.70, N = 3SE +/- 6.46, N = 9SE +/- 6.08, N = 3SE +/- 10.58, N = 311894858779511. (CXX) g++ options: -flto -pthread

libavif avifenc

Encoder Speed: 10, Lossless

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 10, Losslessc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3048121620SE +/- 0.021, N = 3SE +/- 0.021, N = 3SE +/- 0.081, N = 3SE +/- 0.087, N = 35.7657.8777.18413.7091. (CXX) g++ options: -O3 -fPIC -lm

Build2

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3060120180240300SE +/- 0.64, N = 3SE +/- 3.00, N = 12SE +/- 1.62, N = 3SE +/- 1.73, N = 3115.02235.50126.14270.01

TensorFlow Lite

Model: SqueezeNet

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: SqueezeNetc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3016003200480064008000SE +/- 22.07, N = 3SE +/- 54.01, N = 3SE +/- 6.53, N = 3SE +/- 13.10, N = 33257.947508.833665.997230.65

simdjson

Throughput Test: DistinctUserID

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: DistinctUserIDc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-300.811.622.433.244.05SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 32.691.661.873.60-pthread-pthread-pthread1. (CXX) g++ options: -O3

Stress-NG

Test: Memory Copying

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Memory Copyingc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm14002800420056007000SE +/- 3.52, N = 3SE +/- 7.47, N = 3SE +/- 5.84, N = 36693.323189.313740.281. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

Timed Apache Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To Compilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-301326395265SE +/- 0.05, N = 3SE +/- 0.71, N = 3SE +/- 0.03, N = 3SE +/- 0.21, N = 326.9455.9931.3543.90

LeelaChessZero

Backend: BLAS

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302004006008001000SE +/- 6.44, N = 3SE +/- 6.05, N = 4SE +/- 8.59, N = 9SE +/- 8.99, N = 311035559469831. (CXX) g++ options: -flto -pthread

GPAW

Input: Carbon Nanotube

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 22.1Input: Carbon Nanotubec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm70140210280350SE +/- 0.08, N = 3SE +/- 3.47, N = 3SE +/- 2.05, N = 3155.18305.72202.18-pthread-pthread1. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi

DaCapo Benchmark

Java Test: Jython

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Jythonc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3016003200480064008000SE +/- 6.99, N = 4SE +/- 28.16, N = 4SE +/- 17.76, N = 4SE +/- 34.59, N = 43940675348857593

simdjson

Throughput Test: PartialTweets

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: PartialTweetsc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-300.69981.39962.09942.79923.499SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 32.621.651.833.11-pthread-pthread-pthread1. (CXX) g++ options: -O3

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_barec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-300.25380.50760.76141.01521.269SE +/- 0.002, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 31.1280.6100.8580.604-pthread-pthread-pthread1. (CXX) g++ options: -O3

Zstd Compression

Compression Level: 19, Long Mode - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression Speedc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-307001400210028003500SE +/- 6.93, N = 3SE +/- 17.95, N = 3SE +/- 5.72, N = 3SE +/- 27.77, N = 33240.61904.52466.11762.7-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

Zstd Compression

Compression Level: 3 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Decompression Speedc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-308001600240032004000SE +/- 2.07, N = 3SE +/- 9.51, N = 9SE +/- 1.10, N = 2SE +/- 21.89, N = 93508.52095.82825.71954.9-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

Timed MrBayes Analysis

Primate Phylogeny Analysis

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny Analysisc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3070140210280350SE +/- 0.24, N = 3SE +/- 1.14, N = 3SE +/- 1.31, N = 3SE +/- 0.76, N = 3251.40339.72323.12198.15-mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -mabm1. (CC) gcc options: -O3 -std=c99 -pedantic -lm

QuantLib

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21c7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-305001000150020002500SE +/- 0.15, N = 3SE +/- 17.05, N = 15SE +/- 15.63, N = 15SE +/- 18.81, N = 15SE +/- 8.02, N = 32512.71878.41910.91975.51580.51. (CXX) g++ options: -O3 -march=native -rdynamic

libavif avifenc

Encoder Speed: 2

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 2c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3050100150200250SE +/- 0.11, N = 3SE +/- 1.54, N = 3SE +/- 0.20, N = 3SE +/- 0.49, N = 3141.70223.60209.21183.371. (CXX) g++ options: -O3 -fPIC -lm

simdjson

Throughput Test: Kostya

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: Kostyac7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-300.46130.92261.38391.84522.3065SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 81.941.301.472.05-pthread-pthread-pthread1. (CXX) g++ options: -O3

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess Performancec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30300K600K900K1200K1500KSE +/- 0.00, N = 5SE +/- 1332.29, N = 5SE +/- 964.36, N = 5SE +/- 4109.65, N = 51370094104038810407788804761. (CC) gcc options: -O3 -march=native

NAS Parallel Benchmarks

Test / Class: BT.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.Cc7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302K4K6K8K10KSE +/- 7.36, N = 3SE +/- 9.84, N = 3SE +/- 30.27, N = 3SE +/- 3.28, N = 3SE +/- 25.05, N = 310339.536924.506799.067720.309055.91-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

PHPBench

PHP Benchmark Suite

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark Suitec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30140K280K420K560K700KSE +/- 525.83, N = 3SE +/- 1035.45, N = 3SE +/- 1142.02, N = 3SE +/- 3673.12, N = 3666484491649491470445360

Liquid-DSP

Threads: 16 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3080M160M240M320M400MSE +/- 400097.21, N = 3SE +/- 107857.93, N = 3SE +/- 133832.40, N = 3SE +/- 563037.40, N = 33836066673191800003195566672668833331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

NAS Parallel Benchmarks

Test / Class: EP.D

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.Dc7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302004006008001000SE +/- 0.39, N = 3SE +/- 0.34, N = 3SE +/- 1.49, N = 3SE +/- 0.44, N = 3SE +/- 8.74, N = 3934.72661.87660.38671.08713.82-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

ASTC Encoder

Preset: Thorough

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Thoroughc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30510152025SE +/- 0.00, N = 3SE +/- 0.16, N = 9SE +/- 0.14, N = 8SE +/- 0.08, N = 313.9219.0716.8915.671. (CXX) g++ options: -O3 -flto -pthread

Stress-NG

Test: Vector Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Vector Mathc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm12K24K36K48K60KSE +/- 17.05, N = 3SE +/- 7.88, N = 3SE +/- 4.62, N = 355258.1741006.2141030.721. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

Google SynthMark

Test: VoiceMark_100

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30150300450600750SE +/- 0.32, N = 3SE +/- 0.12, N = 3SE +/- 0.22, N = 3SE +/- 5.56, N = 3675.64559.89561.23504.351. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

ASTC Encoder

Preset: Exhaustive

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Exhaustivec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-304080120160200SE +/- 0.01, N = 3SE +/- 0.22, N = 3SE +/- 0.13, N = 3SE +/- 0.37, N = 3139.38183.50156.81140.201. (CXX) g++ options: -O3 -flto -pthread

SecureMark

Benchmark: SecureMark-TLS

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3040K80K120K160K200KSE +/- 773.26, N = 3SE +/- 99.89, N = 3SE +/- 104.19, N = 3SE +/- 1461.11, N = 51837081418781417331472681. (CC) gcc options: -pedantic -O3

simdjson

Throughput Test: LargeRandom

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: LargeRandomc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-300.15750.3150.47250.630.7875SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 150.700.550.570.67-pthread-pthread-pthread1. (CXX) g++ options: -O3

Stress-NG

Test: CPU Stress

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU Stressc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm11002200330044005500SE +/- 0.41, N = 3SE +/- 22.51, N = 3SE +/- 19.08, N = 35029.713974.484016.611. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

WebP Image Encode

Encode Settings: Quality 100, Highest Compression

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Highest Compressionc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-303691215SE +/- 0.007, N = 3SE +/- 0.056, N = 3SE +/- 0.006, N = 3SE +/- 0.140, N = 39.34610.29910.09811.340-pthread -ltiff-pthread -ltiff-pthread -ltiff1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16

PyBench

Total For Average Test Times

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test Timesc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3030060090012001500SE +/- 0.33, N = 3SE +/- 4.48, N = 3SE +/- 7.06, N = 31185142114251413

Stress-NG

Test: Crypto

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Cryptoc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm5K10K15K20K25KSE +/- 32.01, N = 3SE +/- 43.41, N = 3SE +/- 7.66, N = 323181.8120420.0920364.231. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

Stress-NG

Test: Matrix Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Matrix Mathc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm20K40K60K80K100KSE +/- 3.18, N = 3SE +/- 8.70, N = 3SE +/- 28.72, N = 380088.7476853.0176899.281. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

ONNX Runtime

Model: super-resolution-10 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: Standardc7g.4xlarge6001200180024003000SE +/- 1.86, N = 328171. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: ArcFace ResNet-100 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: Standardc7g.4xlarge130260390520650SE +/- 0.00, N = 36091. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: fcn-resnet101-11 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: Standardc7g.4xlarge918273645SE +/- 0.00, N = 3381. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: bertsquad-12 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: Standardc7g.4xlarge90180270360450SE +/- 0.17, N = 34071. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

ONNX Runtime

Model: GPT-2 - Device: CPU - Executor: Standard

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: Standardc7g.4xlarge2K4K6K8K10KSE +/- 2.40, N = 379901. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

Stress-NG

Test: IO_uring

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: IO_uringc7g.4xlarge200K400K600K800K1000KSE +/- 614.16, N = 3843015.781. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

Stress-NG

Test: CPU Cache

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU Cachec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vm50100150200250SE +/- 3.64, N = 12SE +/- 2.46, N = 4SE +/- 0.08, N = 364.31236.50213.471. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

Ngspice

Circuit: C7552

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-302004006008001000SE +/- 1.94, N = 3SE +/- 49.48, N = 9SE +/- 0.31, N = 3SE +/- 0.75, N = 3191.29796.45239.74198.37-lXft -lfontconfig -lXrender -lfreetype-lXft -lfontconfig -lXrender -lfreetype1. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

Ngspice

Circuit: C2670

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30140280420560700SE +/- 0.86, N = 3SE +/- 27.85, N = 6SE +/- 1.01, N = 3SE +/- 0.87, N = 3198.22654.19239.24254.64-lXft -lfontconfig -lXrender -lfreetype-lXft -lfontconfig -lXrender -lfreetype1. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

Timed PHP Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To Compilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-304080120160200SE +/- 0.11, N = 3SE +/- 4.68, N = 12SE +/- 0.10, N = 3SE +/- 0.64, N = 369.48162.6182.04122.81

Timed ImageMagick Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To Compilec7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3020406080100SE +/- 0.13, N = 3SE +/- 3.17, N = 15SE +/- 0.53, N = 3SE +/- 0.29, N = 327.9076.1338.6563.36

libavif avifenc

Encoder Speed: 0

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 0c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3090180270360450SE +/- 0.18, N = 3SE +/- 9.46, N = 9SE +/- 0.65, N = 3SE +/- 1.25, N = 3256.84419.32356.25396.311. (CXX) g++ options: -O3 -fPIC -lm

7-Zip Compression

Test: Compression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 21.06Test: Compression Ratingc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3020K40K60K80K100KSE +/- 159.36, N = 3SE +/- 927.35, N = 12SE +/- 90.01, N = 3SE +/- 137.98, N = 3978244060680614284371. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Zstd Compression

Compression Level: 19 - Decompression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression Speedc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-307001400210028003500SE +/- 7.75, N = 3SE +/- 105.10, N = 15SE +/- 10.05, N = 3SE +/- 15.16, N = 33050.31449.12343.51736.8-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

Zstd Compression

Compression Level: 19 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression Speedc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30918273645SE +/- 0.00, N = 3SE +/- 0.27, N = 15SE +/- 0.03, N = 3SE +/- 0.09, N = 341.214.037.819.0-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

Zstd Compression

Compression Level: 3 - Compression Speed

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression Speedc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3010002000300040005000SE +/- 9.57, N = 3SE +/- 19.16, N = 15SE +/- 11.61, N = 3SE +/- 11.70, N = 94639.11063.32938.61405.5-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

DaCapo Benchmark

Java Test: Tradebeans

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Tradebeansc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-305K10K15K20K25KSE +/- 26.73, N = 4SE +/- 440.17, N = 16SE +/- 23.57, N = 4SE +/- 60.16, N = 2032032443647287572

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP Streamclusterc7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-301224364860SE +/- 0.33, N = 12SE +/- 1.06, N = 15SE +/- 0.09, N = 3SE +/- 0.42, N = 313.3052.3913.3530.891. (CXX) g++ options: -O2 -lOpenCL

NAS Parallel Benchmarks

Test / Class: FT.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.Cc7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-303K6K9K12K15KSE +/- 1.17, N = 3SE +/- 24.16, N = 9SE +/- 47.76, N = 3SE +/- 10.37, N = 3SE +/- 99.16, N = 311791.77650.956338.877344.749878.38-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

NAS Parallel Benchmarks

Test / Class: CG.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.Cc7g.4xlarge16 vcpu ampere Vm16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-3014002800420056007000SE +/- 17.12, N = 3SE +/- 85.05, N = 12SE +/- 101.16, N = 15SE +/- 5.85, N = 3SE +/- 11.73, N = 36571.952651.142436.263826.482958.09-lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz-pthread-pthread-pthread-pthread1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

High Performance Conjugate Gradient

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1c7g.4xlarge16 vcpu ampere Vm run 2pinned 16 vcpu ampere Vmb2-30612182430SE +/- 0.03738, N = 3SE +/- 0.16954, N = 11SE +/- 0.02395, N = 3SE +/- 0.09413, N = 1226.305809.2600317.644905.93594-pthread-pthread-pthread1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Phoronix Test Suite v10.8.4