Amazon EC2 Graviton3 Benchmark Comparison

Amazon AWS Graviton3 benchmarks by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2205260-PTS-GRAVITON42
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts

Limit displaying results to tests within:

BLAS (Basic Linear Algebra Sub-Routine) Tests 2 Tests
C++ Boost Tests 2 Tests
Chess Test Suite 6 Tests
Timed Code Compilation 7 Tests
C/C++ Compiler Tests 15 Tests
Compression Tests 2 Tests
CPU Massive 22 Tests
Creator Workloads 7 Tests
Cryptography 2 Tests
Fortran Tests 4 Tests
Go Language Tests 2 Tests
HPC - High Performance Computing 14 Tests
Imaging 2 Tests
Common Kernel Benchmarks 3 Tests
Linear Algebra 2 Tests
Machine Learning 3 Tests
Molecular Dynamics 4 Tests
MPI Benchmarks 7 Tests
Multi-Core 23 Tests
NVIDIA GPU Compute 3 Tests
OpenMPI Tests 10 Tests
Programmer / Developer System Benchmarks 12 Tests
Python Tests 6 Tests
Raytracing 2 Tests
Renderers 2 Tests
Scientific Computing 8 Tests
Server 5 Tests
Server CPU Tests 16 Tests
Single-Threaded 3 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
a1.4xlarge Graviton
May 25 2022
  17 Hours, 42 Minutes
c6g.4xlarge Graviton2
May 25 2022
  9 Hours, 58 Minutes
c7g.4xlarge Graviton3
May 24 2022
  7 Hours, 43 Minutes
c6a.4xlarge EPYC
May 26 2022
  11 Hours, 43 Minutes
c6i.4xlarge Xeon
May 26 2022
  9 Hours, 40 Minutes
Invert Hiding All Results Option
  11 Hours, 21 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


Amazon EC2 Graviton3 Benchmark ComparisonProcessorMotherboardChipsetMemoryDiskNetworkOSKernelCompilerFile-SystemSystem Layera1.4xlarge Gravitonc6g.4xlarge Graviton2c7g.4xlarge Graviton3c6a.4xlarge EPYCc6i.4xlarge XeonARMv8 Cortex-A72 (16 Cores)Amazon EC2 a1.4xlarge (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)Amazon EC2 c6g.4xlarge (1.0 BIOS)ARMv8 Neoverse-V1 (16 Cores)Amazon EC2 c7g.4xlarge (1.0 BIOS)AMD EPYC 7R13 (8 Cores / 16 Threads)Amazon EC2 c6a.4xlarge (1.0 BIOS)Intel 440FX 82441FX PMC5.15.0-1004-aws (x86_64)Intel Xeon Platinum 8375C (8 Cores / 16 Threads)Amazon EC2 c6i.4xlarge (1.0 BIOS)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- a1.4xlarge Graviton: --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 - c6g.4xlarge Graviton2: --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 - c7g.4xlarge Graviton3: --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 - c6a.4xlarge EPYC: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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 - c6i.4xlarge Xeon: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --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-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-gBFGDP/gcc-11-11.2.0/debian/tmp-gcn/usr --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-build-config=bootstrap-lto-lean --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- OpenJDK Runtime Environment (build 11.0.15+10-Ubuntu-0ubuntu0.22.04.1)Python Details- Python 3.10.4Security Details- a1.4xlarge Graviton: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Branch predictor hardening BHB + srbds: Not affected + tsx_async_abort: Not affected - c6g.4xlarge Graviton2: 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 - c7g.4xlarge Graviton3: 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 - c6a.4xlarge EPYC: 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 Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected - c6i.4xlarge Xeon: 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected Processor Details- c6a.4xlarge EPYC: CPU Microcode: 0xa001144- c6i.4xlarge Xeon: CPU Microcode: 0xd000331

a1.4xlarge Gravitonc6g.4xlarge Graviton2c7g.4xlarge Graviton3c6a.4xlarge EPYCc6i.4xlarge XeonLogarithmic Result OverviewPhoronix Test SuiteHigh Performance Conjugate GradientAlgebraic Multi-Grid BenchmarkACES DGEMMONNX RuntimeXcompact3d Incompact3dNAS Parallel BenchmarksTimed MrBayes AnalysisGPAWLULESHGROMACSApache HTTP ServersimdjsonTensorFlow LiteASTC EncoderTimed Node.js CompilationLAMMPS Molecular Dynamics SimulatorPyBenchPHPBenchlibavif avifencTimed ImageMagick CompilationTimed Apache CompilationTimed LLVM CompilationRodiniaOpenSSLZstd CompressionSecureMarkNgspiceLiquid-DSPBuild2Timed PHP CompilationDaCapo BenchmarkWebP Image EncodeTimed Gem5 CompilationnginxC-RayTSCPStockfishPOV-Ray7-Zip CompressionStress-NGasmFishGoogle SynthMarkLeelaChessZeroCoremarkN-Queensm-queens

Amazon EC2 Graviton3 Benchmark Comparisonstress-ng: Memory Copyingnpb: MG.Cnpb: CG.Cnpb: SP.Ccompress-zstd: 3 - Compression Speednpb: FT.Chpcg: amg: incompact3d: input.i3d 129 Cells Per Directionmt-dgemm: Sustained Floating-Point Rateincompact3d: input.i3d 193 Cells Per Directionstress-ng: CPU Stresssimdjson: DistinctUserIDmrbayes: Primate Phylogeny Analysisnpb: IS.Dtensorflow-lite: Mobilenet Floatgpaw: Carbon Nanotubeavifenc: 2simdjson: PartialTweetslulesh: apache: 100astcenc: Thoroughgromacs: MPI CPU - water_GMX50_baretensorflow-lite: Inception V4apache: 500apache: 200simdjson: Kostyanpb: BT.Copenssl: RSA4096tensorflow-lite: Inception ResNet V2apache: 1000tensorflow-lite: SqueezeNetastcenc: Exhaustiverodinia: OpenMP CFD Solveropenssl: RSA4096avifenc: 0build-nodejs: Time To Compilelammps: Rhodopsin Proteinpybench: Total For Average Test Timesphpbench: PHP Benchmark Suitebuild-imagemagick: Time To Compiletensorflow-lite: NASNet Mobilebuild-apache: Time To Compiledacapobench: Jythonbuild-llvm: Ninjanpb: EP.Dngspice: C2670dacapobench: Tradesoapsimdjson: LargeRandsecuremark: SecureMark-TLSdacapobench: Tradebeansliquid-dsp: 16 - 256 - 57build2: Time To Compilebuild-php: Time To Compilecompress-7zip: Compression Ratingngspice: C7552webp: Quality 100, Lossless, Highest Compressionbuild-gem5: Time To Compilewebp: Quality 100, Losslessavifenc: 6, Losslessnginx: 1000nginx: 500nginx: 200compress-zstd: 19 - Decompression Speednginx: 100tscp: AI Chess Performancecompress-zstd: 19, Long Mode - Decompression Speedstockfish: Total Timerodinia: OpenMP LavaMDpovray: Trace Timecompress-zstd: 19, Long Mode - Compression Speedcompress-zstd: 19 - Compression Speeddacapobench: H2stress-ng: Cryptoasmfish: 1024 Hash Memory, 26 Depthsynthmark: VoiceMark_100openssl: SHA256stress-ng: Vector Mathnpb: LU.Clczero: Eigenlczero: BLAScoremark: CoreMark Size 666 - Iterations Per Secondn-queens: Elapsed Timecompress-7zip: Decompression Ratingm-queens: Time To Solvestress-ng: IO_uringonnx: super-resolution-10 - CPU - Standardonnx: ArcFace ResNet-100 - CPU - Standardonnx: fcn-resnet101-11 - CPU - Standardonnx: bertsquad-12 - CPU - Standardonnx: GPT-2 - CPU - Standardtensorflow-lite: Mobilenet Quantc-ray: Total Time - 4K, 16 Rays Per Pixelrodinia: OpenMP Streamclustera1.4xlarge Gravitonc6g.4xlarge Graviton2c7g.4xlarge Graviton3c6a.4xlarge EPYCc6i.4xlarge Xeon798.243266.361213.151293.80633.92927.163.7783418671693353.77062740.891391182.5839392366.000.8644.788197.579990.15769.346449.0220.782328.272418636.4333.51980.31618891020133.4920887.580.633148.18588.317116919278.6812014.7277.766941.45045328.6768.3021765.9103.245345224125993.63230986.774.742129971784.600339.20473.901111820.3743569045165513333353.912196.02932498480.793124.7081155.61561.80133.991138205.11139414.84141436.201121.7143155.485385001213.910980430360.30493.8011616.9674011985.3815331550331.070678568951727341.472558.12128135203869.40201732.28540891110.368918172.377571651011523125724.66104.76147.4302903.006720.683520.862356.162888.36244.4819.721893265290011.57335474.78512341.02408353404.941.53384.753372.762500.87215.528238.2051.516016.162746995.3516.52220.78146793.950077.8150059.971.196449.11660.645955.746629.453969.35159.203917.03553951.5406.937628.4017.935174144985540.33314985.434.2015626682.981558.88263.72445060.491203014344262890000142.27788.89771285255.20566.147488.80531.08216.518308213.13310596.58308938.672051.6307349.368723132196.321679245215.66651.04731.034.6396417924.1826540482470.3891072318408337753.895133.89834864315464.33980023.1365944575.224770521.8120723342832269481980.2462.32315.4846693.3213481.616571.954467.194639.111791.7726.305812588073338.016714255.85386429.12585705029.712.69251.3971041.902156.60155.180141.6982.6210940.93967231.8813.92481.12841855.173546.3273676.951.9410339.532546.440051.372719.333257.94139.379710.478178460.4256.841497.57911.291118566648427.90411591.926.9403940544.929934.72198.22435240.71837083203383606667115.02069.48397824191.28648.208391.17122.76911.908346814.75346613.34352380.983050.3345710.8713700943240.627608891143.33437.86339.541.2295123181.8132134123675.6351372204597355258.177730.4111891103405413.86055421.5367305466.822843015.7828176093840779901502.9538.51713.2963551.8016826.436169.228094.792784.018299.965.0604226767070028.27976612.432432110.77002713304.504.30120.636541.352159.72302.95693.9463.645452.105177567.697.98181.00444920.681995.6483070.002.8013134.462088.941366.671537.113103.1272.390821.789136784.2195.532664.3475.067196148074132.6269266.8623.5324616760.344466.21245.88640520.952132883167509746667150.99467.08462562180.35648.677515.20126.70816.394388657.76389030.11390932.792907.5388010.7614426312826.023857623224.33149.43525.930.0301913556.0626187688663.0731169140335353787.6125140.5510011091345133.44054116.3785731872.330768723.46369611926548856173847.9669.34918.3833150.4926298.819522.829563.223440.620423.578.6603166136476717.86827722.23054569.216997812527.164.30134.924861.571965.07202.10697.7353.718112.371586545.577.26251.45241185.791746.5794458.222.4613888.402161.341179.779830.962983.9369.638720.446140964.4204.994604.6206.22099782818629.73710900.622.5274013685.7041103.22147.89338150.862305492928373100000136.80164.33766631161.08141.805469.94021.12217.529347345.49351672.92356829.932582.0356302.8412725962666.122081961281.38952.78433.838.1292110210.3423746200565.690709699393740140.3038136.7714661397285378.84166118.8394565391.2311037943.373450137413977379443967.3992.54523.512OpenBenchmarking.org

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Memory Copyinga1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton314002800420056007000SE +/- 0.91, N = 3SE +/- 11.57, N = 3SE +/- 3.75, N = 3SE +/- 0.94, N = 3SE +/- 3.52, N = 3798.243551.802903.003150.496693.321. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.Ca1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton36K12K18K24K30KSE +/- 1.64, N = 3SE +/- 30.62, N = 3SE +/- 1.39, N = 3SE +/- 184.24, N = 3SE +/- 4.69, N = 33266.3616826.436720.6826298.8113481.611. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.Ca1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 11.79, N = 6SE +/- 81.25, N = 3SE +/- 9.95, N = 3SE +/- 66.44, N = 3SE +/- 17.12, N = 31213.156169.223520.869522.826571.951. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.Ca1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 2.51, N = 3SE +/- 24.63, N = 3SE +/- 0.57, N = 3SE +/- 73.65, N = 3SE +/- 9.61, N = 31293.808094.792356.169563.224467.191. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 3 - Compression Speeda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton310002000300040005000SE +/- 4.47, N = 3SE +/- 18.93, N = 3SE +/- 3.74, N = 3SE +/- 29.53, N = 3SE +/- 9.57, N = 3633.92768.72878.83440.64639.1-llzma-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: FT.Ca1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton34K8K12K16K20KSE +/- 1.73, N = 3SE +/- 45.90, N = 3SE +/- 1.10, N = 3SE +/- 40.24, N = 3SE +/- 1.17, N = 32927.1618299.966244.4820423.5711791.771. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

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.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3612182430SE +/- 0.00065, N = 3SE +/- 0.00225, N = 3SE +/- 0.01639, N = 3SE +/- 0.04033, N = 3SE +/- 0.03738, N = 33.778345.0604219.721808.6603126.305801. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3300M600M900M1200M1500MSE +/- 176548.39, N = 3SE +/- 103921.81, N = 3SE +/- 3420043.89, N = 3SE +/- 5114517.12, N = 3SE +/- 952437.28, N = 318671693326767070093265290066136476712588073331. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per Directiona1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton31224364860SE +/- 0.02862870, N = 3SE +/- 0.03718674, N = 3SE +/- 0.01351889, N = 3SE +/- 0.09619197, N = 3SE +/- 0.01401446, N = 353.7706274028.2797661011.5733547017.868277208.016714251. (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

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point Ratea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton31.31712.63423.95135.26846.5855SE +/- 0.002370, N = 3SE +/- 0.023324, N = 6SE +/- 0.007139, N = 3SE +/- 0.003819, N = 3SE +/- 0.016350, N = 30.8913912.4324324.7851232.2305455.8538641. (CC) gcc options: -O3 -march=native -fopenmp

Xcompact3d Incompact3d

Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per Directiona1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton34080120160200SE +/- 0.15, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 3SE +/- 0.14, N = 3SE +/- 0.03, N = 3182.58110.7741.0269.2229.131. (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

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: CPU Stressa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton33K6K9K12K15KSE +/- 0.16, N = 3SE +/- 37.60, N = 3SE +/- 0.54, N = 3SE +/- 155.66, N = 3SE +/- 0.41, N = 32366.0013304.503404.9412527.165029.711. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

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.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: DistinctUserIDa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton30.96751.9352.90253.874.8375SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.804.301.534.302.691. (CXX) g++ options: -O3

Timed MrBayes Analysis

This test performs a bayesian analysis of a set of primate genome sequences in order to estimate their phylogeny. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MrBayes Analysis 3.2.7Primate Phylogeny Analysisa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3140280420560700SE +/- 0.49, N = 3SE +/- 0.35, N = 3SE +/- 0.11, N = 3SE +/- 1.43, N = 3SE +/- 0.24, N = 3644.79120.64384.75134.92251.40-mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4a -msha -maes -mavx -mfma -mavx2 -mrdrnd -mbmi -mbmi2 -madx -mabm-mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msha -maes -mavx -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512bw -mavx512dq -mavx512ifma -mavx512vbmi -mrdrnd -mbmi -mbmi2 -madx -mabm1. (CC) gcc options: -O3 -std=c99 -pedantic -lm

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.Da1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32004006008001000SE +/- 0.31, N = 3SE +/- 0.47, N = 3SE +/- 0.20, N = 3SE +/- 2.14, N = 3SE +/- 2.29, N = 3197.57541.35372.76861.571041.901. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet Floata1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 113.94, N = 3SE +/- 1.03, N = 3SE +/- 28.63, N = 3SE +/- 1.81, N = 3SE +/- 19.61, N = 39990.152159.722500.871965.072156.60

GPAW

GPAW is a density-functional theory (DFT) Python code based on the projector-augmented wave (PAW) method and the atomic simulation environment (ASE). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 22.1Input: Carbon Nanotubea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3170340510680850SE +/- 5.37, N = 3SE +/- 0.17, N = 3SE +/- 0.13, N = 3SE +/- 0.24, N = 3SE +/- 0.08, N = 3769.35302.96215.53202.11155.181. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 2a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3100200300400500SE +/- 0.29, N = 3SE +/- 0.44, N = 3SE +/- 0.12, N = 3SE +/- 0.26, N = 3SE +/- 0.11, N = 3449.0293.95238.2197.74141.701. (CXX) g++ options: -O3 -fPIC -lm

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.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: PartialTweetsa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton30.83481.66962.50443.33924.174SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.783.641.513.712.621. (CXX) g++ options: -O3

LULESH

LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 6.27, N = 3SE +/- 5.52, N = 3SE +/- 4.88, N = 3SE +/- 14.20, N = 3SE +/- 76.73, N = 32328.275452.116016.168112.3710940.941. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 100a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320K40K60K80K100KSE +/- 28.97, N = 3SE +/- 211.56, N = 3SE +/- 93.03, N = 3SE +/- 389.13, N = 3SE +/- 38.09, N = 318636.4377567.6946995.3586545.5767231.881. (CC) gcc options: -shared -fPIC -O2

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Thorougha1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3816243240SE +/- 0.0061, N = 3SE +/- 0.0154, N = 3SE +/- 0.0064, N = 3SE +/- 0.0001, N = 3SE +/- 0.0011, N = 333.51987.981816.52227.262513.92481. (CXX) g++ options: -O3 -flto -pthread

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2022.1Implementation: MPI CPU - Input: water_GMX50_barea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton30.32670.65340.98011.30681.6335SE +/- 0.000, N = 3SE +/- 0.002, N = 3SE +/- 0.001, N = 3SE +/- 0.001, N = 3SE +/- 0.002, N = 30.3161.0040.7811.4521.1281. (CXX) g++ options: -O3

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception V4a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton340K80K120K160K200KSE +/- 1746.17, N = 3SE +/- 53.95, N = 3SE +/- 197.89, N = 3SE +/- 75.14, N = 3SE +/- 210.27, N = 3188910.044920.646793.941185.741855.1

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 500a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320K40K60K80K100KSE +/- 93.64, N = 3SE +/- 636.46, N = 13SE +/- 578.32, N = 3SE +/- 833.50, N = 7SE +/- 89.82, N = 320133.4981995.6450077.8191746.5773546.321. (CC) gcc options: -shared -fPIC -O2

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 200a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320K40K60K80K100KSE +/- 59.55, N = 3SE +/- 644.29, N = 3SE +/- 112.65, N = 3SE +/- 615.05, N = 3SE +/- 649.31, N = 320887.5883070.0050059.9794458.2273676.951. (CC) gcc options: -shared -fPIC -O2

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.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: Kostyaa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton30.631.261.892.523.15SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.632.801.192.461.941. (CXX) g++ options: -O3

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.Ca1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton33K6K9K12K15KSE +/- 3.44, N = 3SE +/- 98.45, N = 3SE +/- 3.20, N = 3SE +/- 22.04, N = 3SE +/- 7.36, N = 33148.1813134.466449.1113888.4010339.531. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton35001000150020002500SE +/- 0.12, N = 3SE +/- 1.40, N = 3SE +/- 0.03, N = 3SE +/- 4.47, N = 3SE +/- 0.23, N = 3588.32088.9660.62161.32546.4-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Inception ResNet V2a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton340K80K120K160K200KSE +/- 825.35, N = 3SE +/- 27.66, N = 3SE +/- 336.95, N = 3SE +/- 110.01, N = 3SE +/- 305.31, N = 3171169.041366.645955.741179.740051.3

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.48Concurrent Requests: 1000a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320K40K60K80K100KSE +/- 98.61, N = 3SE +/- 397.88, N = 3SE +/- 276.10, N = 3SE +/- 335.63, N = 3SE +/- 83.83, N = 319278.6871537.1146629.4579830.9672719.331. (CC) gcc options: -shared -fPIC -O2

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: SqueezeNeta1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton33K6K9K12K15KSE +/- 46.48, N = 3SE +/- 1.37, N = 3SE +/- 37.23, N = 3SE +/- 3.54, N = 3SE +/- 22.07, N = 312014.703103.123969.352983.933257.94

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 3.2Preset: Exhaustivea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton360120180240300SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 3277.7772.39159.2069.64139.381. (CXX) g++ options: -O3 -flto -pthread

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD Solvera1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3918273645SE +/- 0.08, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 341.4521.7917.0420.4510.481. (CXX) g++ options: -O2 -lOpenCL

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.0Algorithm: RSA4096a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton340K80K120K160K200KSE +/- 63.75, N = 3SE +/- 74.60, N = 3SE +/- 3.30, N = 3SE +/- 47.94, N = 3SE +/- 82.61, N = 345328.6136784.253951.5140964.4178460.4-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 0a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3170340510680850SE +/- 0.58, N = 3SE +/- 0.62, N = 3SE +/- 0.13, N = 3SE +/- 0.33, N = 3SE +/- 0.18, N = 3768.30195.53406.94204.99256.841. (CXX) g++ options: -O3 -fPIC -lm

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 17.3Time To Compilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3400800120016002000SE +/- 1.80, N = 3SE +/- 0.26, N = 3SE +/- 0.37, N = 3SE +/- 0.42, N = 3SE +/- 2.06, N = 31765.91664.35628.40604.62497.58

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin Proteina1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton33691215SE +/- 0.040, N = 3SE +/- 0.039, N = 12SE +/- 0.014, N = 3SE +/- 0.009, N = 3SE +/- 0.060, N = 33.2455.0677.9356.22011.2911. (CXX) g++ options: -O3 -lm

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test Timesa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton37001400210028003500SE +/- 18.15, N = 3SE +/- 1.53, N = 3SE +/- 1.67, N = 3SE +/- 3.84, N = 3SE +/- 0.33, N = 33452196117419971185

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark Suitea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3200K400K600K800K1000KSE +/- 816.27, N = 3SE +/- 2681.41, N = 3SE +/- 743.13, N = 3SE +/- 983.65, N = 3SE +/- 525.83, N = 3241259480741449855828186666484

Timed ImageMagick Compilation

This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed ImageMagick Compilation 6.9.0Time To Compilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320406080100SE +/- 0.27, N = 3SE +/- 0.06, N = 3SE +/- 0.22, N = 3SE +/- 0.07, N = 3SE +/- 0.13, N = 393.6332.6340.3329.7427.90

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: NASNet Mobilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton37K14K21K28K35KSE +/- 49.84, N = 3SE +/- 23.44, N = 3SE +/- 203.15, N = 15SE +/- 166.62, N = 14SE +/- 121.56, N = 1530986.709266.8614985.4010900.6011591.90

Timed Apache Compilation

This test times how long it takes to build the Apache HTTPD web server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To Compilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320406080100SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 374.7423.5334.2022.5326.94

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Jythona1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton33K6K9K12K15KSE +/- 48.38, N = 4SE +/- 23.33, N = 4SE +/- 23.29, N = 4SE +/- 24.07, N = 4SE +/- 6.99, N = 4129974616562640133940

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 13.0Build System: Ninjaa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3400800120016002000SE +/- 0.34, N = 3SE +/- 0.05, N = 3SE +/- 0.49, N = 3SE +/- 0.12, N = 3SE +/- 5.19, N = 31784.60760.34682.98685.70544.93

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.Da1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32004006008001000SE +/- 0.24, N = 3SE +/- 0.06, N = 3SE +/- 0.23, N = 3SE +/- 19.93, N = 9SE +/- 0.39, N = 3339.20466.21558.881103.22934.721. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3100200300400500SE +/- 3.48, N = 3SE +/- 1.17, N = 3SE +/- 0.91, N = 3SE +/- 1.80, N = 4SE +/- 0.86, N = 3473.90245.89263.72147.89198.221. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Tradesoapa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 71.92, N = 4SE +/- 16.15, N = 4SE +/- 27.95, N = 4SE +/- 24.39, N = 4SE +/- 14.95, N = 4111824052450638153524

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.orgGB/s, More Is Bettersimdjson 1.0Throughput Test: LargeRandoma1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton30.21380.42760.64140.85521.069SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 30.300.950.490.860.701. (CXX) g++ options: -O3

SecureMark

SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton350K100K150K200K250KSE +/- 59.40, N = 3SE +/- 3310.19, N = 9SE +/- 23.07, N = 3SE +/- 864.34, N = 3SE +/- 773.26, N = 3743562132881203012305491837081. (CC) gcc options: -pedantic -O3

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Tradebeansa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 44.35, N = 4SE +/- 23.53, N = 11SE +/- 40.13, N = 4SE +/- 19.24, N = 20SE +/- 26.73, N = 490453167434429283203

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3110M220M330M440M550MSE +/- 8819.17, N = 3SE +/- 489364.67, N = 3SE +/- 35118.85, N = 3SE +/- 41633.32, N = 3SE +/- 400097.21, N = 31655133335097466672628900003731000003836066671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Build2

This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton380160240320400SE +/- 1.89, N = 3SE +/- 0.87, N = 3SE +/- 0.70, N = 3SE +/- 0.69, N = 3SE +/- 0.64, N = 3353.91150.99142.28136.80115.02

Timed PHP Compilation

This test times how long it takes to build PHP 7. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed PHP Compilation 7.4.2Time To Compilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton34080120160200SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.31, N = 3SE +/- 0.09, N = 3SE +/- 0.11, N = 3196.0367.0888.9064.3469.48

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 21.06Test: Compression Ratinga1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320K40K60K80K100KSE +/- 91.00, N = 3SE +/- 16.02, N = 3SE +/- 44.77, N = 3SE +/- 174.34, N = 3SE +/- 159.36, N = 332498625627128566631978241. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3100200300400500SE +/- 1.19, N = 3SE +/- 0.66, N = 3SE +/- 2.40, N = 7SE +/- 0.33, N = 3SE +/- 1.94, N = 3480.79180.36255.21161.08191.291. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lXft -lfontconfig -lXrender -lfreetype -lSM -lICE

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Lossless, Highest Compressiona1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3306090120150SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.01, N = 3SE +/- 0.34, N = 3SE +/- 0.01, N = 3124.7148.6866.1541.8148.21-ltiff-ltiff-ltiff-ltiff1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16

Timed Gem5 Compilation

This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To Compilea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32004006008001000SE +/- 0.78, N = 3SE +/- 0.79, N = 3SE +/- 0.53, N = 3SE +/- 0.59, N = 3SE +/- 1.33, N = 31155.62515.20488.81469.94391.17

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgEncode Time - Seconds, Fewer Is BetterWebP Image Encode 1.1Encode Settings: Quality 100, Losslessa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton31428425670SE +/- 0.06, N = 3SE +/- 0.17, N = 15SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 361.8026.7131.0821.1222.77-ltiff-ltiff-ltiff-ltiff1. (CC) gcc options: -fvisibility=hidden -O2 -lm -ljpeg -lpng16

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.10Encoder Speed: 6, Losslessa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3816243240SE +/- 0.31, N = 3SE +/- 0.12, N = 3SE +/- 0.17, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 333.9916.3916.5217.5311.911. (CXX) g++ options: -O3 -fPIC -lm

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 1000a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton380K160K240K320K400KSE +/- 66.96, N = 3SE +/- 781.49, N = 3SE +/- 1677.89, N = 3SE +/- 2637.25, N = 3SE +/- 1410.11, N = 3138205.11388657.76308213.13347345.49346814.751. (CC) gcc options: -lcrypt -lz -O3 -march=native

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 500a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton380K160K240K320K400KSE +/- 141.15, N = 3SE +/- 771.95, N = 3SE +/- 3783.68, N = 3SE +/- 1620.39, N = 3SE +/- 1017.52, N = 3139414.84389030.11310596.58351672.92346613.341. (CC) gcc options: -lcrypt -lz -O3 -march=native

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 200a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton380K160K240K320K400KSE +/- 133.96, N = 3SE +/- 1242.81, N = 3SE +/- 1347.28, N = 3SE +/- 1582.66, N = 3SE +/- 3986.77, N = 3141436.20390932.79308938.67356829.93352380.981. (CC) gcc options: -lcrypt -lz -O3 -march=native

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Decompression Speeda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton37001400210028003500SE +/- 4.74, N = 3SE +/- 3.25, N = 3SE +/- 12.10, N = 3SE +/- 24.18, N = 3SE +/- 7.75, N = 31121.72907.52051.62582.03050.3-llzma-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.21.1Concurrent Requests: 100a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton380K160K240K320K400KSE +/- 22.67, N = 3SE +/- 436.72, N = 3SE +/- 3992.58, N = 3SE +/- 1727.81, N = 3SE +/- 2009.97, N = 3143155.48388010.76307349.36356302.84345710.871. (CC) gcc options: -lcrypt -lz -O3 -march=native

TSCP

This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess Performancea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3300K600K900K1200K1500KSE +/- 196.86, N = 5SE +/- 4180.17, N = 5SE +/- 338.27, N = 5SE +/- 1099.67, N = 5SE +/- 0.00, N = 55385001442631872313127259613700941. (CC) gcc options: -O3 -march=native

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Decompression Speeda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton37001400210028003500SE +/- 15.28, N = 3SE +/- 6.53, N = 3SE +/- 2.93, N = 3SE +/- 7.82, N = 3SE +/- 6.93, N = 31213.92826.02196.32666.13240.6-llzma-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 512 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 13Total Timea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton36M12M18M24M30MSE +/- 123749.22, N = 3SE +/- 149731.77, N = 3SE +/- 292329.99, N = 3SE +/- 242448.39, N = 3SE +/- 153578.64, N = 31098043023857623216792452208196127608891-m64 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi2-m64 -msse -msse3 -mpopcnt -mavx2 -mavx512f -mavx512bw -mavx512vnni -mavx512dq -mavx512vl -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

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton380160240320400SE +/- 0.07, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.14, N = 3SE +/- 0.15, N = 3360.30224.33215.67281.39143.331. (CXX) g++ options: -O2 -lOpenCL

POV-Ray

This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace Timea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320406080100SE +/- 0.94, N = 15SE +/- 0.18, N = 3SE +/- 0.00, N = 3SE +/- 0.12, N = 3SE +/- 0.01, N = 393.8049.4451.0552.7837.86-march=native-march=native1. (CXX) g++ options: -pipe -O3 -ffast-math -R/usr/lib -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19, Long Mode - Compression Speeda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3918273645SE +/- 0.00, N = 3SE +/- 0.27, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.23, N = 316.025.931.033.839.5-llzma-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.5.0Compression Level: 19 - Compression Speeda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3918273645SE +/- 0.03, N = 3SE +/- 0.21, N = 3SE +/- 0.06, N = 3SE +/- 0.40, N = 3SE +/- 0.00, N = 316.930.034.638.141.2-llzma-llzma-llzma-llzma1. (CC) gcc options: -O3 -pthread -lz

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton314002800420056007000SE +/- 63.66, N = 4SE +/- 27.42, N = 4SE +/- 45.89, N = 4SE +/- 32.93, N = 4SE +/- 32.57, N = 567403019396429212951

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Cryptoa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton35K10K15K20K25KSE +/- 6.29, N = 3SE +/- 3.93, N = 3SE +/- 92.83, N = 3SE +/- 5.89, N = 3SE +/- 32.01, N = 311985.3813556.0617924.1810210.3423181.811. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

asmFish

This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2018-07-231024 Hash Memory, 26 Deptha1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton37M14M21M28M35MSE +/- 106812.26, N = 3SE +/- 303648.79, N = 3SE +/- 359309.26, N = 3SE +/- 325631.00, N = 3SE +/- 104795.40, N = 31533155026187688265404822374620032134123

Google SynthMark

SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3150300450600750SE +/- 0.00, N = 3SE +/- 7.09, N = 3SE +/- 0.33, N = 3SE +/- 2.00, N = 3SE +/- 0.32, N = 3331.07663.07470.39565.69675.641. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.0Algorithm: SHA256a1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton33000M6000M9000M12000M15000MSE +/- 12563225.46, N = 3SE +/- 8616254.20, N = 3SE +/- 47755430.47, N = 3SE +/- 606684.16, N = 3SE +/- 7739237.92, N = 367856895171169140335310723184083709699393713722045973-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: Vector Matha1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton312K24K36K48K60KSE +/- 0.49, N = 3SE +/- 2.46, N = 3SE +/- 15.72, N = 3SE +/- 28.50, N = 3SE +/- 17.05, N = 327341.4753787.6137753.8940140.3055258.171. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.Ca1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton38K16K24K32K40KSE +/- 0.15, N = 3SE +/- 18.06, N = 3SE +/- 0.90, N = 3SE +/- 160.86, N = 3SE +/- 1.96, N = 32558.1225140.555133.8938136.777730.411. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2

LeelaChessZero

LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: Eigena1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton330060090012001500SE +/- 0.67, N = 3SE +/- 11.74, N = 9SE +/- 12.00, N = 3SE +/- 13.37, N = 3SE +/- 9.70, N = 31281001834146611891. (CXX) g++ options: -flto -pthread

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton330060090012001500SE +/- 0.88, N = 3SE +/- 12.82, N = 9SE +/- 10.22, N = 4SE +/- 12.41, N = 9SE +/- 6.44, N = 31351091864139711031. (CXX) g++ options: -flto -pthread

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Seconda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton390K180K270K360K450KSE +/- 116.54, N = 3SE +/- 2163.00, N = 3SE +/- 49.84, N = 3SE +/- 80.93, N = 3SE +/- 3211.91, N = 3203869.40345133.44315464.34285378.84405413.861. (CC) gcc options: -O2 -lrt" -lrt

N-Queens

This is a test of the OpenMP version of a test that solves the N-queens problem. The board problem size is 18. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed Timea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3714212835SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 332.2916.3823.1418.8421.541. (CC) gcc options: -static -fopenmp -O3 -march=native

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 21.06Test: Decompression Ratinga1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton316K32K48K64K80KSE +/- 31.21, N = 3SE +/- 142.56, N = 3SE +/- 239.68, N = 3SE +/- 35.00, N = 3SE +/- 12.88, N = 340891573185944545653730541. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

m-queens

A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.2Time To Solvea1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320406080100SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.00, N = 3110.3772.3375.2291.2366.821. (CXX) g++ options: -fopenmp -O2 -march=native

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14Test: IO_uringa1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3200K400K600K800K1000KSE +/- 3840.04, N = 3SE +/- 713.03, N = 3SE +/- 2395.13, N = 3SE +/- 405.56, N = 3SE +/- 614.16, N = 3918172.37768723.46770521.811037943.37843015.781. (CC) gcc options: -O2 -std=gnu99 -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lz -pthread

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: super-resolution-10 - Device: CPU - Executor: Standarda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton38001600240032004000SE +/- 0.50, N = 3SE +/- 234.97, N = 12SE +/- 1.74, N = 3SE +/- 1.61, N = 3SE +/- 1.86, N = 375736962072345028171. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: ArcFace ResNet-100 - Device: CPU - Executor: Standarda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton330060090012001500SE +/- 0.50, N = 3SE +/- 82.60, N = 12SE +/- 0.17, N = 3SE +/- 91.51, N = 12SE +/- 0.00, N = 3165119233413746091. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: fcn-resnet101-11 - Device: CPU - Executor: Standarda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3306090120150SE +/- 0.00, N = 3SE +/- 5.55, N = 12SE +/- 0.00, N = 3SE +/- 0.60, N = 3SE +/- 0.00, N = 3106528139381. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: bertsquad-12 - Device: CPU - Executor: Standarda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton3170340510680850SE +/- 0.88, N = 3SE +/- 0.58, N = 3SE +/- 0.17, N = 3SE +/- 50.92, N = 12SE +/- 0.17, N = 31154883227734071. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.11Model: GPT-2 - Device: CPU - Executor: Standarda1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton32K4K6K8K10KSE +/- 2.20, N = 3SE +/- 75.29, N = 12SE +/- 3.50, N = 3SE +/- 322.41, N = 12SE +/- 2.40, N = 3231256176948794479901. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation focused on TensorFlow machine learning for mobile, IoT, edge, and other cases. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2022-05-18Model: Mobilenet Quanta1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton312002400360048006000SE +/- 20.90, N = 3SE +/- 53.31, N = 15SE +/- 14.44, N = 3SE +/- 80.05, N = 12SE +/- 17.76, N = 35724.663847.961980.243967.391502.95

C-Ray

This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per Pixela1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton320406080100SE +/- 2.00, N = 15SE +/- 0.77, N = 5SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3104.7669.3562.3292.5538.521. (CC) gcc options: -lm -lpthread -O3

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP Streamclustera1.4xlarge Gravitonc6a.4xlarge EPYCc6g.4xlarge Graviton2c6i.4xlarge Xeonc7g.4xlarge Graviton31122334455SE +/- 0.02, N = 3SE +/- 0.05, N = 3SE +/- 0.26, N = 15SE +/- 0.07, N = 3SE +/- 0.33, N = 1247.4318.3815.4823.5113.301. (CXX) g++ options: -O2 -lOpenCL

94 Results Shown

Stress-NG
NAS Parallel Benchmarks:
  MG.C
  CG.C
  SP.C
Zstd Compression
NAS Parallel Benchmarks
High Performance Conjugate Gradient
Algebraic Multi-Grid Benchmark
Xcompact3d Incompact3d
ACES DGEMM
Xcompact3d Incompact3d
Stress-NG
simdjson
Timed MrBayes Analysis
NAS Parallel Benchmarks
TensorFlow Lite
GPAW
libavif avifenc
simdjson
LULESH
Apache HTTP Server
ASTC Encoder
GROMACS
TensorFlow Lite
Apache HTTP Server:
  500
  200
simdjson
NAS Parallel Benchmarks
OpenSSL
TensorFlow Lite
Apache HTTP Server
TensorFlow Lite
ASTC Encoder
Rodinia
OpenSSL
libavif avifenc
Timed Node.js Compilation
LAMMPS Molecular Dynamics Simulator
PyBench
PHPBench
Timed ImageMagick Compilation
TensorFlow Lite
Timed Apache Compilation
DaCapo Benchmark
Timed LLVM Compilation
NAS Parallel Benchmarks
Ngspice
DaCapo Benchmark
simdjson
SecureMark
DaCapo Benchmark
Liquid-DSP
Build2
Timed PHP Compilation
7-Zip Compression
Ngspice
WebP Image Encode
Timed Gem5 Compilation
WebP Image Encode
libavif avifenc
nginx:
  1000
  500
  200
Zstd Compression
nginx
TSCP
Zstd Compression
Stockfish
Rodinia
POV-Ray
Zstd Compression:
  19, Long Mode - Compression Speed
  19 - Compression Speed
DaCapo Benchmark
Stress-NG
asmFish
Google SynthMark
OpenSSL
Stress-NG
NAS Parallel Benchmarks
LeelaChessZero:
  Eigen
  BLAS
Coremark
N-Queens
7-Zip Compression
m-queens
Stress-NG
ONNX Runtime:
  super-resolution-10 - CPU - Standard
  ArcFace ResNet-100 - CPU - Standard
  fcn-resnet101-11 - CPU - Standard
  bertsquad-12 - CPU - Standard
  GPT-2 - CPU - Standard
TensorFlow Lite
C-Ray
Rodinia