AmpereOne A192-32X vs. AWS Graviton4 CPU Performance Benchmarks

AmpereOne versus AWS Graviton4 ARM64 CPU benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2409052-NE-GRAVITON412
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
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
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
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
AmpereOne A192-32X
August 25
  11 Hours, 34 Minutes
Graviton4 192 vCPUs
September 05
  7 Hours, 20 Minutes
Invert Behavior (Only Show Selected Data)
  9 Hours, 27 Minutes
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):

AmpereOne A192-32X

Processor: AmpereOne @ 3.20GHz (192 Cores), Motherboard: Supermicro ARS-211M-NR R13SPD v1.02 (T20240726102529 BIOS), Chipset: Ampere Computing LLC Device e208, Memory: 8 x 64GB DDR5-5200MT/s, Disk: 3841GB SAMSUNG MZQL23T8HCLS-00A07 + 960GB SAMSUNG MZ1L2960HCJR-00A07, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Broadcom BCM57414 NetXtreme-E 10Gb/25Gb + 2 x Mellanox MT2892

OS: Ubuntu 24.04, Kernel: 6.8.0-39-generic-64k (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Graviton4 192 vCPUs

Processor: ARMv8 Neoverse-V2 (192 Cores), Motherboard: Amazon EC2 r8g.48xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 1520GB, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-41-generic-64k (aarch64), Compiler: GCC 13.2.0, File-System: ext4, System Layer: amazon

Kernel Notes: nvme_core.io_timeout=4294967295 - Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

AmpereOne A192-32X vs. AWS Graviton4 CPU Performance BenchmarksProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelCompilerFile-SystemScreen ResolutionSystem LayerAmpereOne A192-32XGraviton4 192 vCPUsAmpereOne @ 3.20GHz (192 Cores)Supermicro ARS-211M-NR R13SPD v1.02 (T20240726102529 BIOS)Ampere Computing LLC Device e2088 x 64GB DDR5-5200MT/s3841GB SAMSUNG MZQL23T8HCLS-00A07 + 960GB SAMSUNG MZ1L2960HCJR-00A07ASPEEDVGA HDMI2 x Broadcom BCM57414 NetXtreme-E 10Gb/25Gb + 2 x Mellanox MT2892Ubuntu 24.046.8.0-39-generic-64k (aarch64)GCC 13.2.0ext41920x1080ARMv8 Neoverse-V2 (192 Cores)Amazon EC2 r8g.48xlarge (1.0 BIOS)Amazon Device 02001520GB429GB Amazon Elastic Block StoreAmazon Elastic6.8.0-41-generic-64k (aarch64)amazonOpenBenchmarking.orgKernel Details- AmpereOne A192-32X: Transparent Huge Pages: always- Graviton4 192 vCPUs: nvme_core.io_timeout=4294967295 - Transparent Huge Pages: madviseCompiler Details- --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v Processor Details- AmpereOne A192-32X: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)Python Details- Python 3.12.3Security Details- AmpereOne A192-32X: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected - Graviton4 192 vCPUs: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

AmpereOne A192-32X vs. Graviton4 192 vCPUs ComparisonPhoronix Test SuiteBaseline+61.2%+61.2%+122.4%+122.4%+183.6%+183.6%244.6%218.4%215.9%208.5%186.1%173%170%157.2%154.8%146.3%145.7%142.1%110.1%93.2%89.3%88.9%88.1%82.7%82.1%81.9%80.5%78.9%78.2%77.8%77%74.2%73.9%68.1%66.5%65.3%64.2%56.3%55.4%47.5%46.3%45.5%44.6%43.7%43.3%43.1%42.1%38.9%35.2%34.7%33.9%28.1%24.8%23.4%22.8%15.1%13.6%11.1%9.4%8%7.9%7.7%5.6%144 144 144 - 60X.b.i.ii.i.1.C.P.Dd.M.M.S - Execution TimetConvolve MPI - GriddingtConvolve MPI - DegriddingleblancbigsedovbigNoise-Gaussianconus 2.5km128 - 256 - 32EnhancedVery ThoroughThoroughExhaustived.M.M.S - Mesh Time1.R.H.D.S.RMulti-ThreadedMPI CPU - water_GMX50_bareChess Benchmark1.R.H.D.T.Rclover_bm64_short1.R.H.D.F.R.C.CSharpenIS.DRand ReadRand ReadSwirld.S.M.S - Mesh TimeNinjaRhodopsin ProteinT.F.A.T.TSP.CC240 BuckyballLi2_STO_aeCarbon Nanotubed.S.M.S - Execution Time20k AtomsSmallEP.DTime To CompileTime To CompileCompression RatingBlowfish26.5%bcrypt24.9%Time To CompileP.P.B.T.TP.P.B.T.T14 digit100 - 1000 - Read Only14.6%100 - 1000 - Read Only - Average Latency14.6%P.P.B.T.TCoreMark Size 666 - I.P.SGhostRider - 1M1:100clover_bm16F.1.0.R.a.m.i.CTime To Solve7.1%P.P.B.T.T1e135.3%High Performance Conjugate GradientAlgebraic Multi-Grid BenchmarkXcompact3d Incompact3dXcompact3d Incompact3dOpenFOAMLULESHASKAPASKAPPennantPennantGraphicsMagickWRFLiquid-DSPGraphicsMagickASTC EncoderASTC EncoderASTC EncoderOpenFOAMClickHouseQuantLibGROMACSStockfishClickHouseCloverLeafClickHouseGraphicsMagickNAS Parallel BenchmarksRocksDBSpeedbGraphicsMagickOpenFOAMTimed LLVM CompilationLAMMPS Molecular Dynamics SimulatorPyBenchNAS Parallel BenchmarksNWChemQMCPACKGPAWNumpy BenchmarkOpenFOAMLAMMPS Molecular Dynamics SimulatorminiFENAS Parallel BenchmarksTimed Gem5 CompilationTimed Node.js Compilation7-Zip CompressionJohn The RipperJohn The RipperTimed Mesa CompilationsrsRAN ProjectsrsRAN ProjectHelsingPostgreSQLPostgreSQLsrsRAN ProjectCoremarkXmrigMemcachedCloverLeafParallel BZIP2 Compressionm-queenssrsRAN ProjectPrimesieveAmpereOne A192-32XGraviton4 192 vCPUs

AmpereOne A192-32X vs. AWS Graviton4 CPU Performance Benchmarkswrf: conus 2.5kmclickhouse: 100M Rows Hits Dataset, Third Runclickhouse: 100M Rows Hits Dataset, Second Runclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachehpcg: 144 144 144 - 60nwchem: C240 Buckyballstockfish: Chess Benchmarkpgbench: 100 - 1000 - Read Only - Average Latencypgbench: 100 - 1000 - Read Onlycloverleaf: clover_bm16memcached: 1:100incompact3d: X3D-benchmarking input.i3dnumpy: build-nodejs: Time To Compilerocksdb: Read While Writingbuild-gem5: Time To Compilexmrig: GhostRider - 1Mgraphics-magick: Noise-Gaussianbuild-llvm: Ninjalammps: 20k Atomsopenfoam: drivaerFastback, Medium Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timegpaw: Carbon Nanotubeqmcpack: Li2_STO_aecloverleaf: clover_bm64_shortaskap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddingspeedb: Rand Readgraphics-magick: Sharpengraphics-magick: Enhancedrocksdb: Rand Readgraphics-magick: Swirlminife: Smallpytorch: CPU - 512 - ResNet-50srsran: PUSCH Processor Benchmark, Throughput Totalsrsran: PUSCH Processor Benchmark, Throughput Threadquantlib: Multi-Threadedamg: npb: SP.Cgromacs: MPI CPU - water_GMX50_barejohn-the-ripper: Blowfishjohn-the-ripper: bcryptliquid-dsp: 128 - 256 - 32astcenc: Very Thoroughastcenc: Exhaustivehelsing: 14 digitcompress-7zip: Decompression Ratingcompress-7zip: Compression Ratingpybench: Total For Average Test Timespennant: sedovbigopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Small Mesh Size - Mesh Timecoremark: CoreMark Size 666 - Iterations Per Secondm-queens: Time To Solvepennant: leblancbigincompact3d: input.i3d 193 Cells Per Directionsrsran: PDSCH Processor Benchmark, Throughput Totalnpb: EP.Dsrsran: PDSCH Processor Benchmark, Throughput Threadprimesieve: 1e13lulesh: build-mesa: Time To Compilenpb: IS.Dastcenc: Thoroughlammps: Rhodopsin Proteincompress-pbzip2: FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionAmpereOne A192-32XGraviton4 192 vCPUs9102.318406.36400.27402.8733.36172293.21286021360.3642746730349.493895708.19290.280721313.43214.1559803710195.56817812.6245177.10453.859315.65587152.175441.060106.9139.2318414.720809.273440163546937071920337195338349.219.982824.750.2300839.9182826533333612.417.03917630617404829839000007.37484.593030.47387730575668112464.29058632.35887128.4174554561177.3361705.3573.3494398.9558148424257.27557.72193.714.01041890.40717.1922417.0450.023250.5351.6444853759.19724.05728.98713.11114.9511576.32301115880.4172396138323.864205545.7291.8868612449.08159.9669720737145.16919482.7602113.27776.529110.3330683.28068928.58373.95922.0649717.853520.612226182718177151208755766157553285.03208.553547281.9582099233349186.0212.708139322139374626833333313.96158.637226.4778742389690918451.74203222.6186917.3024845066982.4997025.7401.3143592.9028760329796.910218.17239.114.746114373.3513.7774203.9594.486278.5531.526932OpenBenchmarking.org

WRF

WRF, the Weather Research and Forecasting Model, is a "next-generation mesoscale numerical weather prediction system designed for both atmospheric research and operational forecasting applications. It features two dynamical cores, a data assimilation system, and a software architecture supporting parallel computation and system extensibility." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWRF 4.2.2Input: conus 2.5kmGraviton4 192 vCPUsAmpereOne A192-32X2K4K6K8K10K3759.199102.321. (F9X) gfortran options: -O2 -ftree-vectorize -funroll-loops -ffree-form -fconvert=big-endian -frecord-marker=4 -fallow-invalid-boz -lesmf_time -lwrfio_nf -lnetcdff -lnetcdf -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ / https://github.com/ClickHouse/ClickBench/tree/main/clickhouse with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all separate queries performed as an aggregate. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunGraviton4 192 vCPUsAmpereOne A192-32X160320480640800SE +/- 6.40, N = 3SE +/- 13.28, N = 9724.05406.36MIN: 90.09 / MAX: 6000MIN: 30.14 / MAX: 5000

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunGraviton4 192 vCPUsAmpereOne A192-32X160320480640800SE +/- 7.44, N = 3SE +/- 15.66, N = 9728.98400.27MIN: 89.42 / MAX: 6000MIN: 30.88 / MAX: 5000

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheGraviton4 192 vCPUsAmpereOne A192-32X150300450600750SE +/- 1.29, N = 3SE +/- 17.49, N = 9713.11402.87MIN: 90.5 / MAX: 5454.55MIN: 23.07 / MAX: 5000

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 3.1X Y Z: 144 144 144 - RT: 60Graviton4 192 vCPUsAmpereOne A192-32X306090120150SE +/- 0.04, N = 3SE +/- 0.39, N = 3114.9533.361. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 BuckyballGraviton4 192 vCPUsAmpereOne A192-32X50010001500200025001576.32293.21. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -lcomex -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

Stockfish

This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 1024 CPU threads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 16.1Chess BenchmarkGraviton4 192 vCPUsAmpereOne A192-32X50M100M150M200M250MSE +/- 4060681.61, N = 15SE +/- 4643374.23, N = 112301115881286021361. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -funroll-loops -flto -flto-partition=one -flto=jobserver

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read Only - Average LatencyAmpereOne A192-32XGraviton4 192 vCPUs0.09380.18760.28140.37520.469SE +/- 0.003, N = 12SE +/- 0.005, N = 40.3640.4171. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 16Scaling Factor: 100 - Clients: 1000 - Mode: Read OnlyAmpereOne A192-32XGraviton4 192 vCPUs600K1200K1800K2400K3000KSE +/- 20807.81, N = 12SE +/- 28406.87, N = 4274673023961381. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm16Graviton4 192 vCPUsAmpereOne A192-32X80160240320400SE +/- 0.91, N = 3SE +/- 0.18, N = 3323.86349.491. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100Graviton4 192 vCPUsAmpereOne A192-32X900K1800K2700K3600K4500KSE +/- 241067.66, N = 13SE +/- 76864.47, N = 124205545.723895708.191. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

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: X3D-benchmarking input.i3dGraviton4 192 vCPUsAmpereOne A192-32X60120180240300SE +/- 0.06, N = 3SE +/- 0.24, N = 391.89290.281. (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

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkGraviton4 192 vCPUsAmpereOne A192-32X100200300400500SE +/- 0.13, N = 3SE +/- 0.70, N = 3449.08313.43

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 21.7.2Time To CompileGraviton4 192 vCPUsAmpereOne A192-32X50100150200250SE +/- 0.49, N = 3SE +/- 0.34, N = 3159.97214.16

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While WritingAmpereOne A192-32XGraviton4 192 vCPUs2M4M6M8M10MSE +/- 27064.97, N = 3SE +/- 332194.39, N = 15980371097207371. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

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 23.0.1Time To CompileGraviton4 192 vCPUsAmpereOne A192-32X4080120160200SE +/- 0.73, N = 3SE +/- 0.58, N = 3145.17195.57

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MGraviton4 192 vCPUsAmpereOne A192-32X4K8K12K16K20KSE +/- 112.11, N = 3SE +/- 274.17, N = 1519482.717812.61. (CXX) g++ options: -fexceptions -fno-rtti -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: Noise-GaussianGraviton4 192 vCPUsAmpereOne A192-32X130260390520650SE +/- 8.81, N = 12SE +/- 1.20, N = 3602245-ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lzstd -llzma1. (CC) gcc options: -fopenmp -O2 -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lbz2 -lz -lm -lpthread -lgomp

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaGraviton4 192 vCPUsAmpereOne A192-32X4080120160200SE +/- 0.18, N = 3SE +/- 0.57, N = 3113.28177.10

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 23Jun2022Model: 20k AtomsGraviton4 192 vCPUsAmpereOne A192-32X20406080100SE +/- 0.04, N = 3SE +/- 0.04, N = 376.5353.861. (CXX) g++ options: -O3 -lm -ldl

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution TimeGraviton4 192 vCPUsAmpereOne A192-32X70140210280350110.33315.661. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeGraviton4 192 vCPUsAmpereOne A192-32X30609012015083.28152.181. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

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 23.6Input: Carbon NanotubeGraviton4 192 vCPUsAmpereOne A192-32X918273645SE +/- 0.26, N = 15SE +/- 0.10, N = 328.5841.061. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.17.1Input: Li2_STO_aeGraviton4 192 vCPUsAmpereOne A192-32X20406080100SE +/- 0.20, N = 3SE +/- 0.15, N = 373.96106.911. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -mcpu=native -O3 -lm -ldl

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeaf 1.3Input: clover_bm64_shortGraviton4 192 vCPUsAmpereOne A192-32X918273645SE +/- 0.23, N = 15SE +/- 0.13, N = 322.0639.231. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingGraviton4 192 vCPUsAmpereOne A192-32X11K22K33K44K55KSE +/- 131.17, N = 3SE +/- 93.27, N = 349717.818414.71. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - DegriddingGraviton4 192 vCPUsAmpereOne A192-32X11K22K33K44K55KSE +/- 152.03, N = 3SE +/- 159.94, N = 353520.620809.21. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Speedb

Speedb is a next-generation key value storage engine that is RocksDB compatible and aiming for stability, efficiency, and performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random ReadGraviton4 192 vCPUsAmpereOne A192-32X300M600M900M1200M1500MSE +/- 420037.99, N = 3SE +/- 1533105.14, N = 312226182717344016351. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SharpenGraviton4 192 vCPUsAmpereOne A192-32X2004006008001000SE +/- 4.36, N = 3SE +/- 4.33, N = 3817469-ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lzstd -llzma1. (CC) gcc options: -fopenmp -O2 -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lbz2 -lz -lm -lpthread -lgomp

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: EnhancedGraviton4 192 vCPUsAmpereOne A192-32X150300450600750SE +/- 1.53, N = 3SE +/- 0.33, N = 3715370-ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lzstd -llzma1. (CC) gcc options: -fopenmp -O2 -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lbz2 -lz -lm -lpthread -lgomp

RocksDB

This is a benchmark of Meta/Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Random ReadGraviton4 192 vCPUsAmpereOne A192-32X300M600M900M1200M1500MSE +/- 683858.75, N = 3SE +/- 904748.20, N = 312087557667192033711. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

GraphicsMagick

This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample high resolution (currently 15400 x 6940) JPEG image. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.43Operation: SwirlGraviton4 192 vCPUsAmpereOne A192-32X30060090012001500SE +/- 6.77, N = 3SE +/- 2.52, N = 31575953-ltiff -ljbig -lsharpyuv -lwebp -lwebpmux -lzstd -llzma1. (CC) gcc options: -fopenmp -O2 -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -lxml2 -lbz2 -lz -lm -lpthread -lgomp

miniFE

MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCG Mflops, More Is BetterminiFE 2.2Problem Size: SmallGraviton4 192 vCPUsAmpereOne A192-32X11K22K33K44K55KSE +/- 1848.80, N = 15SE +/- 628.57, N = 1553285.038349.21. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi

PyTorch

This is a benchmark of PyTorch making use of pytorch-benchmark [https://github.com/LukasHedegaard/pytorch-benchmark]. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbatches/sec, More Is BetterPyTorch 2.2.1Device: CPU - Batch Size: 512 - Model: ResNet-50AmpereOne A192-32X510152025SE +/- 0.20, N = 319.98MIN: 12.43 / MAX: 21.16

Device: CPU - Batch Size: 512 - Model: ResNet-50

Graviton4 192 vCPUs: The test quit with a non-zero exit status. E: AttributeError: 'NoneType' object has no attribute 'max'

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput TotalGraviton4 192 vCPUsAmpereOne A192-32X7001400210028003500SE +/- 0.20, N = 3SE +/- 0.12, N = 33208.52824.7MIN: 1852.1 / MAX: 3208.9MIN: 1697.4 / MAX: 2824.91. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PUSCH Processor Benchmark, Throughput ThreadGraviton4 192 vCPUsAmpereOne A192-32X1224364860SE +/- 0.00, N = 3SE +/- 0.00, N = 353.050.2MIN: 35.3MIN: 31.71. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.32Configuration: Multi-ThreadedGraviton4 192 vCPUsAmpereOne A192-32X120K240K360K480K600KSE +/- 302.91, N = 3SE +/- 97.87, N = 3547281.9300839.91. (CXX) g++ options: -O3 -march=native -fPIE -pie

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 1.2Graviton4 192 vCPUsAmpereOne A192-32X1200M2400M3600M4800M6000MSE +/- 2157483.59, N = 3SE +/- 1007826.76, N = 3582099233318282653331. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

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: SP.CGraviton4 192 vCPUsAmpereOne A192-32X11K22K33K44K55KSE +/- 94.96, N = 3SE +/- 8.41, N = 349186.0233612.411. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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 2024Implementation: MPI CPU - Input: water_GMX50_bareGraviton4 192 vCPUsAmpereOne A192-32X3691215SE +/- 0.002, N = 3SE +/- 0.007, N = 312.7087.0391. (CXX) g++ options: -O3 -lm

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: BlowfishAmpereOne A192-32XGraviton4 192 vCPUs40K80K120K160K200KSE +/- 564.17, N = 3SE +/- 351.56, N = 31763061393221. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt -lbz2

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptAmpereOne A192-32XGraviton4 192 vCPUs40K80K120K160K200KSE +/- 1839.06, N = 3SE +/- 201.45, N = 31740481393741. (CC) gcc options: -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt -lbz2

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 1.6Threads: 128 - Buffer Length: 256 - Filter Length: 32Graviton4 192 vCPUsAmpereOne A192-32X1300M2600M3900M5200M6500MSE +/- 1545243.60, N = 3SE +/- 100000.00, N = 3626833333329839000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

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.orgMT/s, More Is BetterASTC Encoder 4.7Preset: Very ThoroughGraviton4 192 vCPUsAmpereOne A192-32X48121620SE +/- 0.0026, N = 3SE +/- 0.0004, N = 313.96157.37481. (CXX) g++ options: -O3 -flto -pthread

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ExhaustiveGraviton4 192 vCPUsAmpereOne A192-32X246810SE +/- 0.0006, N = 3SE +/- 0.0011, N = 38.63724.59301. (CXX) g++ options: -O3 -flto -pthread

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 14 digitGraviton4 192 vCPUsAmpereOne A192-32X714212835SE +/- 0.01, N = 3SE +/- 0.11, N = 326.4830.471. (CC) gcc options: -O2 -pthread

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 22.01Test: Decompression RatingAmpereOne A192-32XGraviton4 192 vCPUs200K400K600K800K1000KSE +/- 2055.40, N = 3SE +/- 24922.61, N = 38773058742381. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression RatingGraviton4 192 vCPUsAmpereOne A192-32X200K400K600K800K1000KSE +/- 4521.10, N = 3SE +/- 7822.06, N = 39690917566811. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

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 TimesGraviton4 192 vCPUsAmpereOne A192-32X30060090012001500SE +/- 0.00, N = 3SE +/- 0.33, N = 38451246

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigGraviton4 192 vCPUsAmpereOne A192-32X0.96541.93082.89623.86164.827SE +/- 0.012223, N = 3SE +/- 0.065187, N = 151.7420324.2905861. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution TimeGraviton4 192 vCPUsAmpereOne A192-32X81624324022.6232.361. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeGraviton4 192 vCPUsAmpereOne A192-32X71421283517.3028.421. (CXX) g++ options: -std=c++14 -O3 -mcpu=native -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

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 SecondGraviton4 192 vCPUsAmpereOne A192-32X1.1M2.2M3.3M4.4M5.5MSE +/- 61511.29, N = 4SE +/- 3500.33, N = 35066982.504561177.341. (CC) gcc options: -O2 -lrt" -lrt

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 SolveAmpereOne A192-32XGraviton4 192 vCPUs1.29152.5833.87455.1666.4575SE +/- 0.010, N = 7SE +/- 0.102, N = 155.3575.7401. (CXX) g++ options: -fopenmp -O2 -march=native

Pennant

Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigGraviton4 192 vCPUsAmpereOne A192-32X0.75361.50722.26083.01443.768SE +/- 0.001158, N = 3SE +/- 0.062865, N = 151.3143593.3494391. (CXX) g++ options: -fopenmp -lmpi_cxx -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 193 Cells Per DirectionGraviton4 192 vCPUsAmpereOne A192-32X3691215SE +/- 0.02410444, N = 9SE +/- 0.04916545, N = 42.902876038.955814841. (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

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput TotalGraviton4 192 vCPUsAmpereOne A192-32X6K12K18K24K30KSE +/- 334.78, N = 3SE +/- 45.30, N = 329796.924257.21. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

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.DGraviton4 192 vCPUsAmpereOne A192-32X2K4K6K8K10KSE +/- 30.70, N = 3SE +/- 27.71, N = 310218.177557.721. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

srsRAN Project

srsRAN Project is a complete ORAN-native 5G RAN solution created by Software Radio Systems (SRS). The srsRAN Project radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.10.1-20240325Test: PDSCH Processor Benchmark, Throughput ThreadGraviton4 192 vCPUsAmpereOne A192-32X50100150200250SE +/- 0.06, N = 3SE +/- 0.24, N = 4239.1193.71. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -ldl

Primesieve

Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve primarily benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 12.1Length: 1e13AmpereOne A192-32XGraviton4 192 vCPUs48121620SE +/- 0.06, N = 4SE +/- 0.01, N = 314.0114.751. (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 2.0.3Graviton4 192 vCPUsAmpereOne A192-32X20K40K60K80K100KSE +/- 74.63, N = 3SE +/- 79.76, N = 3114373.3541890.411. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

Timed Mesa Compilation

This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Mesa Compilation 24.0Time To CompileGraviton4 192 vCPUsAmpereOne A192-32X48121620SE +/- 0.12, N = 3SE +/- 0.07, N = 313.7817.19

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.DGraviton4 192 vCPUsAmpereOne A192-32X9001800270036004500SE +/- 8.10, N = 3SE +/- 10.73, N = 34203.952417.041. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz

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.orgMT/s, More Is BetterASTC Encoder 4.7Preset: ThoroughGraviton4 192 vCPUsAmpereOne A192-32X20406080100SE +/- 0.03, N = 3SE +/- 0.01, N = 494.4950.021. (CXX) g++ options: -O3 -flto -pthread

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 23Jun2022Model: Rhodopsin ProteinGraviton4 192 vCPUsAmpereOne A192-32X20406080100SE +/- 0.56, N = 3SE +/- 3.06, N = 1278.5550.541. (CXX) g++ options: -O3 -lm -ldl

Parallel BZIP2 Compression

This test measures the time needed to compress a file (FreeBSD-13.0-RELEASE-amd64-memstick.img) using Parallel BZIP2 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParallel BZIP2 Compression 1.1.13FreeBSD-13.0-RELEASE-amd64-memstick.img CompressionGraviton4 192 vCPUsAmpereOne A192-32X0.370.741.111.481.85SE +/- 0.418215, N = 15SE +/- 0.008273, N = 111.5269321.6444851. (CXX) g++ options: -O2 -pthread -lbz2 -lpthread

AmpereOne A192-32X

Processor: AmpereOne @ 3.20GHz (192 Cores), Motherboard: Supermicro ARS-211M-NR R13SPD v1.02 (T20240726102529 BIOS), Chipset: Ampere Computing LLC Device e208, Memory: 8 x 64GB DDR5-5200MT/s, Disk: 3841GB SAMSUNG MZQL23T8HCLS-00A07 + 960GB SAMSUNG MZ1L2960HCJR-00A07, Graphics: ASPEED, Monitor: VGA HDMI, Network: 2 x Broadcom BCM57414 NetXtreme-E 10Gb/25Gb + 2 x Mellanox MT2892

OS: Ubuntu 24.04, Kernel: 6.8.0-39-generic-64k (aarch64), Compiler: GCC 13.2.0, File-System: ext4, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Processor Notes: Scaling Governor: cppc_cpufreq performance (Boost: Disabled)
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Not affected + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 25 August 2024 00:56 by user ubuntu.

Graviton4 192 vCPUs

Processor: ARMv8 Neoverse-V2 (192 Cores), Motherboard: Amazon EC2 r8g.48xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 1520GB, Disk: 429GB Amazon Elastic Block Store, Network: Amazon Elastic

OS: Ubuntu 24.04, Kernel: 6.8.0-41-generic-64k (aarch64), Compiler: GCC 13.2.0, File-System: ext4, System Layer: amazon

Kernel Notes: nvme_core.io_timeout=4294967295 - Transparent Huge Pages: madvise
Compiler Notes: --build=aarch64-linux-gnu --disable-libquadmath --disable-libquadmath-support --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-fix-cortex-a53-843419 --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-13-dIwDw0/gcc-13-13.2.0/debian/tmp-nvptx/usr --enable-plugin --enable-shared --enable-threads=posix --host=aarch64-linux-gnu --program-prefix=aarch64-linux-gnu- --target=aarch64-linux-gnu --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-target-system-zlib=auto --without-cuda-driver -v
Python Notes: Python 3.12.3
Security Notes: gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 5 September 2024 15:35 by user ubuntu.