Amazon AWS Graviton3E vs. Graviton 2/3 benchmarks

Benchmarks by Michael Larabel for a future article on Phoronix.com.

HTML result view exported from: https://openbenchmarking.org/result/2405291-NE-2308110NE13&sro&grr.

Amazon AWS Graviton3E vs. Graviton 2/3 benchmarksProcessorMotherboardChipsetMemoryDiskNetworkGraphicsAudioMonitorOSKernelCompilerFile-SystemSystem LayerVulkanDisplay ServerDisplay DriverOpenCLScreen Resolutionm7g.16xlarge Graviton3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Ec6a.16xlarge AMD Zen 3egeo-07ARMv8 Neoverse-V1 (64 Cores)Amazon EC2 m7g.16xlarge (1.0 BIOS)Amazon Device 0200256GB215GB Amazon Elastic Block StoreAmazon ElasticUbuntu 22.045.19.0-1025-aws (aarch64)GCC 11.3.0ext4amazonARMv8 Neoverse-N1 (64 Cores)Amazon EC2 c6g.16xlarge (1.0 BIOS)128GBARMv8 Neoverse-V1 (64 Cores)Amazon EC2 c7g.16xlarge (1.0 BIOS)Amazon EC2 c7gn.16xlarge (1.0 BIOS)AMD EPYC 7R13 (32 Cores / 64 Threads)Amazon EC2 c6a.16xlarge (1.0 BIOS)Intel 440FX 82441FX PMC322GB Amazon Elastic Block Store5.19.0-1025-aws (x86_64)1.3.238GCC 11.4.02 x Intel Xeon Silver 4208 @ 3.20GHz (16 Cores / 32 Threads)Dell Precision 7920 Rack 0DY2X0 (2.21.2 BIOS)Intel Sky Lake-E DMI3 Registers64GB2000GB TOSHIBA DT01ACA2Matrox G200eW3 15GBNVIDIA TU104 HD AudioDELL 17FP4 x Intel I350Debian 115.10.0-28-amd64 (x86_64)X ServerNVIDIAOpenCL 3.0 CUDA 12.2.1381.3.242GCC 10.2.1 20210110 + Clang 11.0.1-2 + CUDA 11.21280x1024OpenBenchmarking.orgKernel Details- m7g.16xlarge Graviton3: Transparent Huge Pages: madvise- c6g.16xlarge Graviton2: Transparent Huge Pages: madvise- c7g.16xlarge Graviton3: Transparent Huge Pages: madvise- c7gn.16xlarge Graviton3E: Transparent Huge Pages: madvise- c6a.16xlarge AMD Zen 3: Transparent Huge Pages: madvise- egeo-07: Transparent Huge Pages: alwaysCompiler Details- m7g.16xlarge 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 - c6g.16xlarge 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.16xlarge 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 - c7gn.16xlarge Graviton3E: --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.16xlarge AMD Zen 3: --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-XeT9lY/gcc-11-11.4.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-XeT9lY/gcc-11-11.4.0/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 - egeo-07: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --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-mutex --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-Km9U7s/gcc-10-10.2.1/debian/tmp-gcn/usr,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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 Python Details- m7g.16xlarge Graviton3: Python 3.10.6- c6g.16xlarge Graviton2: Python 3.10.6- c7g.16xlarge Graviton3: Python 3.10.6- c7gn.16xlarge Graviton3E: Python 3.10.6- c6a.16xlarge AMD Zen 3: Python 3.10.12- egeo-07: Python 2.7.18 + Python 3.9.2Security Details- m7g.16xlarge Graviton3: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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- c6g.16xlarge Graviton2: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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.16xlarge Graviton3: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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- c7gn.16xlarge Graviton3E: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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.16xlarge AMD Zen 3: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: conditional RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected - egeo-07: gather_data_sampling: Mitigation of Microcode + itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Mitigation of Clear buffers; SMT vulnerable + retbleed: Mitigation of Enhanced IBRS + spec_rstack_overflow: 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 PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled Processor Details- c6a.16xlarge AMD Zen 3: CPU Microcode: 0xa0011cf- egeo-07: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x5003605

Amazon AWS Graviton3E vs. Graviton 2/3 benchmarksnwchem: C240 Buckyballgraph500: 26graph500: 26graph500: 26graph500: 26lammps: 20k Atomsbuild-gem5: Time To Compilebrl-cad: VGR Performance Metricstockfish: Total Timelczero: BLASbuild-nodejs: Time To Compilelczero: Eigenqmcpack: FeCO6_b3lyp_gmsqmcpack: FeCO6_b3lyp_gmsbuild-godot: Time To Compilemocassin: Dust 2D tau100.0openssl: SHA256openssl: AES-128-GCMopenssl: ChaCha20openssl: ChaCha20-Poly1305openssl: AES-256-GCMopenssl: SHA512qmcpack: Li2_STO_aestress-ng: CPU Cachelaghos: Sedov Blast Wave, ube_922_hex.meshnekrs: TurboPipe Periodicmt-dgemm: Sustained Floating-Point Ratenpb: EP.Dgpaw: Carbon Nanotubenekrs: Kershawnpb: SP.Cnginx: 1000nginx: 500stress-ng: Wide Vector Mathlaghos: Triple Point Problemrodinia: OpenMP LavaMDheffte: c2c - FFTW - double - 512gromacs: MPI CPU - water_GMX50_barenpb: LU.Copenssl: RSA4096openssl: RSA4096coremark: CoreMark Size 666 - Iterations Per Secondrodinia: OpenMP Streamclusterstress-ng: NUMAqmcpack: simple-H2Osrsran: PUSCH Processor Benchmark, Throughput Totalstress-ng: Vector Floating Pointsrsran: Downlink Processor Benchmarksrsran: PUSCH Processor Benchmark, Throughput Threadheffte: r2c - FFTW - double - 512heffte: c2c - FFTW - float - 512kripke: compress-7zip: Decompression Ratingcompress-7zip: Compression Ratingincompact3d: input.i3d 193 Cells Per Directionliquid-dsp: 64 - 256 - 512liquid-dsp: 32 - 256 - 512stress-ng: Fused Multiply-Addstress-ng: Vector Shuffleliquid-dsp: 64 - 256 - 32liquid-dsp: 64 - 256 - 57stress-ng: Matrix 3D Mathstress-ng: Matrix Mathstress-ng: Memory Copyingliquid-dsp: 32 - 256 - 32liquid-dsp: 32 - 256 - 57stress-ng: Vector Mathremhos: Sample Remap Examplepennant: sedovbigamg: heffte: r2c - FFTW - float - 512mocassin: Gas HII40lulesh: pennant: leblancbigincompact3d: input.i3d 129 Cells Per Directionheffte: c2c - FFTW - double - 256npb: CG.Crodinia: OpenMP CFD Solverheffte: c2c - FFTW - float - 256heffte: r2c - FFTW - double - 256npb: MG.Cheffte: r2c - FFTW - float - 256lammps: Rhodopsin Proteinheffte: c2c - FFTW - double - 128heffte: r2c - FFTW - float - 128heffte: c2c - FFTW - float - 128heffte: r2c - FFTW - double - 128m7g.16xlarge Graviton3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Ec6a.16xlarge AMD Zen 3egeo-071940.24197540002994970001227790000119432000036.927180.2477837771121197111301237.7831398211.60205.72154.37882.669542125155803320331719001032267845177428746099028333311363032125448870112.613892396.34410.55397630000024.3623533738.9861.831315068000017244.85255616.04255768.441542834.94232.0143.78846.25044.22328341.68713859.510181.91601880.34226411.6633759.1028.0415413.876102.55318.595.884.473988.048233900040028554031682513.94541801627533338139666763762252.7654143.402270500000144240000010403.93368750.6720484.241136066667721493333217235.5914.0409.2064901646761667162.95613.57528296.3786.7205373.0987103840.892321988.994.37581.444278.504950126.29164.87337.55857.1503306.540186.356138.0142976.928468900020935000087438900086043200025.171225.30553302086609284947287.814891302.19297.94218.276145.37442472798847158436163857672925412034671763680712919959315714393925490165.121921785.20322.37222019000020.4179522216.2692.76017603366679711.70158676.40148964.69997272.65180.8062.22424.26582.76718741.90214040.92624.31260642.17702413.7352112.6645.2253938.742850.82197.263.844.929742.828422012023323420224070225.88256581349266676748633337732190.5435614.5115314000009782000005752.17284713.6311324.79765466667489270000147886.1420.74016.48050103558633381.941220.75817557.48512.176835.6372073520.627913103.626.05141.981640.110425671.2992.399625.95032.7468209.496135.35881.44981962.74157580002938260001206990000117771000036.862181.7797890661173164761333238.5431382211.32204.77156.68782.822542165612633320643498431032755169977431884221328337379573732145914147112.643844101.98408.01397898333324.1406053664.5462.083326185333317219.95255552.05255145.521535336.57230.6843.96346.37064.20028375.71713945.910181.41605948.67464511.6253523.5827.9905356.876178.46319.795.784.745188.184235444273328563331105613.83266931627666678141200063818458.6154472.072271966667144236666710813.59368671.3920478.671136133333721386667217446.1214.1209.4222701765277667163.27613.65928708.6566.9613453.1444799940.828321911.024.44281.009677.768549742.30162.01037.41255.1055301.418184.026133.51419144117620002961640001207760000117564000036.838182.4717447431170271211392238.6361444188.28204.25155.95182.974541542185934111304699431141181194237996946548735115246542032126059040113.203860335.38423.11414144000024.0785293657.6756.440330282333317163.11256585.83253518.511530043.52236.2244.04446.53004.82028369.11713754.810183.31611801.55926510.6903525.1727.9995431.276911.74323.297.485.006088.455135423406728567731200913.76067261627566678139400063723431.5554695.042266833333144266666710882.02369258.8920475.961136000000721380000217567.1014.0829.3409531765966333163.55913.52528736.2266.8399983.1148982840.970822155.364.42981.167178.165849860.68162.36137.48255.1038300.396184.110133.4223440.420455000015768800041777700041057100020.342192.118485038969056091316230.4231152184.10187.32147.737194.435458575347771514492693171383893787539252299937313845788945015291283297123.951447265.35275.9243375366679.3880503061.4289.818430881000034025.35163178.67165847.751380146.63227.4064.17923.52123.96595221.40548396.58392.41466587.0365808.396552.6826.8676479.196529.51691.3215.942.439444.317623708765023578723097030.314528846007666727480333330920910.9222255.84218486666717108000004571.96147576.418080.4311939666671444266667221776.1522.10416.5305083699930082.758412.66916708.2589.9175657.0197528820.871920210.009.34243.590741.586845946.81102.65219.56348.9432158.85898.702686.37301098485755800690706002105720002082880007.539462.59910268426092344641.867608.86552.72391.737281.2493466925203611313202535621349462726507041907446191285603878566000408.451525522.4670.772.0942061329.20257.46411726.5570348.7773654.19399851.8561.58212.54510.66531.11636556.34200342.73041.8362563.28837023.2470.8977.5231129.521347.32342.388.518.990219.6715109107233590557584084.664220212956000012825000010084119.217162.566412433334778966671841.5855962.003209.8363348333342959666738515.8968.95689.0049044445613334.367426.9205676.776243.4588021.23927379.323799661.9118.63016.997117.117019477.2332.42877.1769.6229658.201226.736723.0160OpenBenchmarking.org

NWChem

Input: C240 Buckyball

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 Buckyballc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton32K4K6K8K10K3440.42976.91962.71914.010984.01940.2-m64-ldl -lutil -m641. (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

Graph500

Scale: 26

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton390M180M270M360M450M20455000028468900041575800041176200085755800419754000-pthread1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Graph500

Scale: 26

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton360M120M180M240M300M15768800020935000029382600029616400069070600299497000-pthread1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Graph500

Scale: 26

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3300M600M900M1200M1500M4177770008743890001206990000120776000021057200012277900001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

Graph500

Scale: 26

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3300M600M900M1200M1500M4105710008604320001177710000117564000020828800011943200001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

LAMMPS Molecular Dynamics Simulator

Model: 20k Atoms

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: 20k Atomsc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3816243240SE +/- 0.066, N = 3SE +/- 0.009, N = 3SE +/- 0.025, N = 3SE +/- 0.018, N = 3SE +/- 0.006, N = 3SE +/- 0.034, N = 320.34225.17136.86236.8387.53936.927-lm-pthread -lm1. (CXX) g++ options: -O3 -ldl

Timed Gem5 Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To Compilec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3100200300400500SE +/- 0.26, N = 3SE +/- 0.35, N = 3SE +/- 0.26, N = 3SE +/- 0.38, N = 3SE +/- 9.98, N = 9SE +/- 0.13, N = 3192.12225.31181.78182.47462.60180.25

BRL-CAD

VGR Performance Metric

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.34VGR Performance Metricc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3200K400K600K800K1000K485038533020789066744743102684783777-m641. (CXX) g++ options: -std=c++14 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -ltcl8.6 -lregex_brl -lz_brl -lnetpbm -ldl -lm -ltk8.6

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 15Total Timec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton330M60M90M120M150MSE +/- 1430593.84, N = 15SE +/- 2597495.37, N = 15SE +/- 2998209.87, N = 12SE +/- 1531345.46, N = 15SE +/- 349749.32, N = 12SE +/- 2854071.93, N = 15969056098660928411731647611702712126092344112119711-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 -fno-peel-loops -fno-tracer -pedantic -O3 -flto -flto=jobserver

LeelaChessZero

Backend: BLAS

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: BLASc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Em7g.16xlarge Graviton330060090012001500SE +/- 13.29, N = 5SE +/- 11.79, N = 3SE +/- 3.53, N = 3SE +/- 7.22, N = 3SE +/- 4.67, N = 313169471333139213011. (CXX) g++ options: -flto -pthread

Timed Node.js Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To Compilec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3140280420560700SE +/- 0.40, N = 3SE +/- 0.16, N = 3SE +/- 0.20, N = 3SE +/- 0.32, N = 3SE +/- 1.11, N = 3SE +/- 0.33, N = 3230.42287.81238.54238.64641.87237.78

LeelaChessZero

Backend: Eigen

OpenBenchmarking.orgNodes Per Second, More Is BetterLeelaChessZero 0.28Backend: Eigenc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Em7g.16xlarge Graviton330060090012001500SE +/- 7.37, N = 3SE +/- 4.73, N = 3SE +/- 15.65, N = 3SE +/- 14.88, N = 3SE +/- 8.74, N = 311528911382144413981. (CXX) g++ options: -flto -pthread

QMCPACK

Input: FeCO6_b3lyp_gms

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.16Input: FeCO6_b3lyp_gmsc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3130260390520650SE +/- 1.03, N = 3SE +/- 0.37, N = 3SE +/- 0.19, N = 3SE +/- 0.29, N = 3SE +/- 0.11, N = 3SE +/- 0.22, N = 3184.10302.19211.32188.28608.86211.60-march=native-mcpu=native-mcpu=native-mcpu=native-march=native -pthread-mcpu=native1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl

QMCPACK

Input: FeCO6_b3lyp_gms

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.16Input: FeCO6_b3lyp_gmsc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3120240360480600SE +/- 2.30, N = 3SE +/- 1.75, N = 3SE +/- 0.82, N = 3SE +/- 0.21, N = 3SE +/- 7.86, N = 3SE +/- 0.45, N = 3187.32297.94204.77204.25552.72205.72-march=native-mcpu=native-mcpu=native-mcpu=native-march=native -pthread-mcpu=native1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl

Timed Godot Game Engine Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To Compilec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton380160240320400SE +/- 0.12, N = 3SE +/- 0.30, N = 3SE +/- 0.63, N = 3SE +/- 0.45, N = 3SE +/- 0.36, N = 3SE +/- 0.32, N = 3147.74218.28156.69155.95391.74154.38

Monte Carlo Simulations of Ionised Nebulae

Input: Dust 2D tau100.0

OpenBenchmarking.orgSeconds, Fewer Is BetterMonte Carlo Simulations of Ionised Nebulae 2.02.73.3Input: Dust 2D tau100.0c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton360120180240300SE +/- 1.84, N = 7SE +/- 0.86, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.37, N = 3SE +/- 0.01, N = 3194.44145.3782.8282.97281.2582.67-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O2 -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lz

OpenSSL

Algorithm: SHA256

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA256c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton312000M24000M36000M48000M60000MSE +/- 26770675.21, N = 3SE +/- 245440310.03, N = 3SE +/- 16491036.11, N = 3SE +/- 19542665.92, N = 3SE +/- 404619.57, N = 3SE +/- 18610524.10, N = 345857534777424727988475421656126354154218593346692520354212515580-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenSSL

Algorithm: AES-128-GCM

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-128-GCMc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton390000M180000M270000M360000M450000MSE +/- 4227452.23, N = 3SE +/- 9833681.11, N = 3SE +/- 12264074.61, N = 3SE +/- 11273100.69, N = 3SE +/- 11737066.92, N = 3SE +/- 81289574.27, N = 315144926931715843616385733206434984341113046994361131320253332033171900-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenSSL

Algorithm: ChaCha20

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton330000M60000M90000M120000M150000MSE +/- 36376378.52, N = 3SE +/- 35952887.59, N = 3SE +/- 1725060.95, N = 3SE +/- 771581.87, N = 3SE +/- 13595278.49, N = 3SE +/- 1293723.80, N = 31383893787536729254120310327551699711411811942356213494627103226784517-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenSSL

Algorithm: ChaCha20-Poly1305

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20-Poly1305c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320000M40000M60000M80000M100000MSE +/- 232372675.93, N = 3SE +/- 1132293.08, N = 3SE +/- 1218886.42, N = 3SE +/- 1769561.47, N = 3SE +/- 1523000.86, N = 3SE +/- 1340503.89, N = 3925229993734671763680774318842213799694654872650704190774287460990-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenSSL

Algorithm: AES-256-GCM

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-256-GCMc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton380000M160000M240000M320000M400000MSE +/- 41584947.90, N = 3SE +/- 2312792.64, N = 3SE +/- 33807617.40, N = 3SE +/- 24279491.44, N = 3SE +/- 2585526.42, N = 3SE +/- 6411836.47, N = 313845788945012919959315728337379573735115246542044619128560283333113630-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenSSL

Algorithm: SHA512

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton37000M14000M21000M28000M35000MSE +/- 207279.55, N = 3SE +/- 9173912.49, N = 3SE +/- 4573992.60, N = 3SE +/- 16155877.53, N = 3SE +/- 1513929.31, N = 3SE +/- 17714077.14, N = 315291283297143939254903214591414732126059040387856600032125448870-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

QMCPACK

Input: Li2_STO_ae

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.16Input: Li2_STO_aec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton390180270360450SE +/- 0.13, N = 3SE +/- 1.13, N = 3SE +/- 0.12, N = 3SE +/- 0.31, N = 3SE +/- 4.45, N = 3SE +/- 0.08, N = 3123.95165.12112.64113.20408.45112.61-march=native-mcpu=native-mcpu=native-mcpu=native-march=native -pthread-mcpu=native1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl

Stress-NG

Test: CPU Cache

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: CPU Cachec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3800K1600K2400K3200K4000KSE +/- 30785.49, N = 12SE +/- 21905.72, N = 15SE +/- 59376.56, N = 15SE +/- 40698.46, N = 15SE +/- 22640.51, N = 15SE +/- 57217.78, N = 151447265.351921785.203844101.983860335.381525522.463892396.34-laio -lbsd -lEGL -lGLESv2 -lmd1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Laghos

Test: Sedov Blast Wave, ube_922_hex.mesh

OpenBenchmarking.orgMajor Kernels Total Rate, More Is BetterLaghos 3.1Test: Sedov Blast Wave, ube_922_hex.meshc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton390180270360450SE +/- 0.48, N = 3SE +/- 0.89, N = 3SE +/- 0.89, N = 3SE +/- 0.79, N = 3SE +/- 0.27, N = 3SE +/- 0.42, N = 3275.92322.37408.01423.1170.77410.55-pthread1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

nekRS

Input: TurboPipe Periodic

OpenBenchmarking.orgflops/rank, More Is BetternekRS 23.0Input: TurboPipe Periodicc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Em7g.16xlarge Graviton3900M1800M2700M3600M4500MSE +/- 12801180.07, N = 3SE +/- 144222.05, N = 3SE +/- 169148.19, N = 3SE +/- 1394740.12, N = 3SE +/- 1199180.28, N = 3433753666722201900003978983333414144000039763000001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi

ACES DGEMM

Sustained Floating-Point Rate

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point Ratec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3612182430SE +/- 0.038051, N = 3SE +/- 0.154503, N = 3SE +/- 0.285590, N = 4SE +/- 0.297525, N = 4SE +/- 0.035680, N = 15SE +/- 0.171001, N = 139.38805020.41795224.14060524.0785292.09420624.3623531. (CC) gcc options: -O3 -march=native -fopenmp

NAS Parallel Benchmarks

Test / Class: EP.D

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.Dc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton38001600240032004000SE +/- 4.77, N = 3SE +/- 2.22, N = 3SE +/- 34.07, N = 15SE +/- 32.06, N = 15SE +/- 0.38, N = 3SE +/- 1.69, N = 33061.422216.263664.543657.671329.203738.98-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. c6a.16xlarge AMD Zen 3: Open MPI 4.1.23. egeo-07: Open MPI 4.1.0

GPAW

Input: Carbon Nanotube

OpenBenchmarking.orgSeconds, Fewer Is BetterGPAW 23.6Input: Carbon Nanotubec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton360120180240300SE +/- 0.13, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.19, N = 3SE +/- 0.03, N = 389.8292.7662.0856.44257.4661.83-pthread1. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi

nekRS

Input: Kershaw

OpenBenchmarking.orgflops/rank, More Is BetternekRS 23.0Input: Kershawc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Em7g.16xlarge Graviton3900M1800M2700M3600M4500MSE +/- 22342148.51, N = 3SE +/- 737119.02, N = 3SE +/- 2490845.46, N = 3SE +/- 5414395.42, N = 3SE +/- 1575066.14, N = 3430881000017603366673261853333330282333331506800001. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -lmpi

NAS Parallel Benchmarks

Test / Class: SP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.Cc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton37K14K21K28K35KSE +/- 20.85, N = 3SE +/- 1.54, N = 3SE +/- 7.21, N = 3SE +/- 31.31, N = 3SE +/- 15.52, N = 3SE +/- 10.19, N = 334025.359711.7017219.9517163.1111726.5517244.85-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. c6a.16xlarge AMD Zen 3: Open MPI 4.1.23. egeo-07: Open MPI 4.1.0

nginx

Connections: 1000

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton350K100K150K200K250KSE +/- 136.82, N = 3SE +/- 185.79, N = 3SE +/- 55.97, N = 3SE +/- 402.16, N = 3SE +/- 141.39, N = 3SE +/- 137.20, N = 3163178.67158676.40255552.05256585.8370348.77255616.041. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

nginx

Connections: 500

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton350K100K150K200K250KSE +/- 60.38, N = 3SE +/- 90.87, N = 3SE +/- 243.69, N = 3SE +/- 317.05, N = 3SE +/- 47.73, N = 3SE +/- 323.56, N = 3165847.75148964.69255145.52253518.5173654.19255768.441. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Stress-NG

Test: Wide Vector Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Wide Vector Mathc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3300K600K900K1200K1500KSE +/- 2507.18, N = 3SE +/- 505.84, N = 3SE +/- 16521.46, N = 15SE +/- 16444.95, N = 15SE +/- 641.34, N = 3SE +/- 16116.93, N = 151380146.63997272.651535336.571530043.52399851.851542834.94-laio -lbsd -lEGL -lGLESv2 -lmd1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Laghos

Test: Triple Point Problem

OpenBenchmarking.orgMajor Kernels Total Rate, More Is BetterLaghos 3.1Test: Triple Point Problemc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton350100150200250SE +/- 1.06, N = 3SE +/- 0.48, N = 3SE +/- 0.16, N = 3SE +/- 0.27, N = 3SE +/- 0.71, N = 4SE +/- 0.28, N = 3227.40180.80230.68236.2261.58232.01-pthread1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

Rodinia

Test: OpenMP LavaMD

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LavaMDc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton350100150200250SE +/- 0.53, N = 3SE +/- 0.04, N = 3SE +/- 0.11, N = 3SE +/- 0.15, N = 3SE +/- 0.02, N = 3SE +/- 0.15, N = 364.1862.2243.9644.04212.5543.79-O2 -lOpenCL-O2 -lOpenCL-O2 -lOpenCL-O2 -lOpenCL-m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl-O2 -lOpenCL1. (CXX) g++ options:

HeFFTe - Highly Efficient FFT for Exascale

Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton31122334455SE +/- 0.05, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 323.5224.2746.3746.5310.6746.25-pthread1. (CXX) g++ options: -O3

GROMACS

Implementation: MPI CPU - Input: water_GMX50_bare

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2023Implementation: MPI CPU - Input: water_GMX50_barec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton31.08452.1693.25354.3385.4225SE +/- 0.013, N = 3SE +/- 0.002, N = 3SE +/- 0.004, N = 3SE +/- 0.003, N = 3SE +/- 0.002, N = 3SE +/- 0.003, N = 33.9652.7674.2004.8201.1164.223-lm1. (CXX) g++ options: -O3

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.Cc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320K40K60K80K100KSE +/- 90.22, N = 3SE +/- 26.12, N = 3SE +/- 36.09, N = 3SE +/- 43.73, N = 3SE +/- 21.23, N = 3SE +/- 48.62, N = 395221.4018741.9028375.7128369.1136556.3428341.68-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. c6a.16xlarge AMD Zen 3: Open MPI 4.1.23. egeo-07: Open MPI 4.1.0

OpenSSL

Algorithm: RSA4096

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.1Algorithm: RSA4096c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3150K300K450K600K750KSE +/- 34.73, N = 3SE +/- 88.30, N = 3SE +/- 12.03, N = 3SE +/- 198.10, N = 3SE +/- 170.27, N = 3SE +/- 21.82, N = 3548396.5214040.9713945.9713754.8200342.7713859.5-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

OpenSSL

Algorithm: RSA4096

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.1Algorithm: RSA4096c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton32K4K6K8K10KSE +/- 3.06, N = 3SE +/- 1.71, N = 3SE +/- 1.54, N = 3SE +/- 0.84, N = 3SE +/- 5.58, N = 3SE +/- 1.27, N = 38392.42624.310181.410183.33041.810181.9-m64-m641. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl

Coremark

CoreMark Size 666 - Iterations Per Second

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per Secondc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3300K600K900K1200K1500KSE +/- 6710.50, N = 3SE +/- 153.60, N = 3SE +/- 13274.76, N = 15SE +/- 14869.41, N = 7SE +/- 4039.35, N = 3SE +/- 11449.37, N = 151466587.041260642.181605948.671611801.56362563.291601880.341. (CC) gcc options: -O2 -lrt" -lrt

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP Streamclusterc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3612182430SE +/- 0.101, N = 15SE +/- 0.211, N = 15SE +/- 0.099, N = 8SE +/- 0.233, N = 12SE +/- 0.397, N = 15SE +/- 0.138, N = 38.39613.73511.62510.69023.24711.663-O2 -lOpenCL-O2 -lOpenCL-O2 -lOpenCL-O2 -lOpenCL-m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl-O2 -lOpenCL1. (CXX) g++ options:

Stress-NG

Test: NUMA

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: NUMAc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton38001600240032004000SE +/- 9.75, N = 15SE +/- 1.53, N = 3SE +/- 3.39, N = 3SE +/- 7.31, N = 3SE +/- 0.00, N = 3SE +/- 5.17, N = 3552.682112.663523.583525.170.893759.101. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

QMCPACK

Input: simple-H2O

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.16Input: simple-H2Oc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.08, N = 3SE +/- 0.24, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.86, N = 5SE +/- 0.03, N = 326.8745.2327.9928.0077.5228.04-march=native-mcpu=native-mcpu=native-mcpu=native-march=native -pthread-mcpu=native1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Total

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.5Test: PUSCH Processor Benchmark, Throughput Totalc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton314002800420056007000SE +/- 21.76, N = 3SE +/- 2.53, N = 3SE +/- 1.80, N = 3SE +/- 3.32, N = 3SE +/- 6.96, N = 3SE +/- 4.08, N = 36479.13938.75356.85431.21129.55413.8-march=native -mfma-march=native -mfma -lpthread1. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -lgtest

Stress-NG

Test: Vector Floating Point

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Floating Pointc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320K40K60K80K100KSE +/- 864.23, N = 13SE +/- 31.31, N = 3SE +/- 71.97, N = 3SE +/- 1.74, N = 3SE +/- 160.53, N = 3SE +/- 190.19, N = 396529.5142850.8276178.4676911.7421347.3276102.55-laio -lbsd -lEGL -lGLESv2 -lmd1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

srsRAN Project

Test: Downlink Processor Benchmark

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.5Test: Downlink Processor Benchmarkc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3150300450600750SE +/- 1.26, N = 3SE +/- 0.25, N = 3SE +/- 0.95, N = 3SE +/- 0.06, N = 3SE +/- 4.00, N = 4SE +/- 0.91, N = 3691.3197.2319.7323.2342.3318.5-march=native -mfma-march=native -mfma -lpthread1. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -lgtest

srsRAN Project

Test: PUSCH Processor Benchmark, Throughput Thread

OpenBenchmarking.orgMbps, More Is BettersrsRAN Project 23.5Test: PUSCH Processor Benchmark, Throughput Threadc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton350100150200250SE +/- 0.55, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.68, N = 10SE +/- 0.03, N = 3215.963.895.797.488.595.8-march=native -mfma-march=native -mfma -lpthread1. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -lgtest

HeFFTe - Highly Efficient FFT for Exascale

Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.02, N = 342.4444.9384.7585.0118.9984.47-pthread1. (CXX) g++ options: -O3

HeFFTe - Highly Efficient FFT for Exascale

Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.14, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.02, N = 344.3242.8388.1888.4619.6788.05-pthread1. (CXX) g++ options: -O3

Kripke

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.6c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton380M160M240M320M400MSE +/- 2932840.19, N = 4SE +/- 102787.75, N = 3SE +/- 525406.56, N = 3SE +/- 445212.18, N = 3SE +/- 523405.33, N = 3SE +/- 619419.33, N = 3237087650220120233354442733354234067109107233339000400-pthread1. (CXX) g++ options: -O3 -fopenmp -ldl

7-Zip Compression

Test: Decompression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression Ratingc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton360K120K180K240K300KSE +/- 1190.65, N = 3SE +/- 15.43, N = 3SE +/- 146.43, N = 3SE +/- 54.90, N = 3SE +/- 265.06, N = 3SE +/- 93.51, N = 3235787234202285633285677590552855401. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

7-Zip Compression

Test: Compression Rating

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression Ratingc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton370K140K210K280K350KSE +/- 670.46, N = 3SE +/- 209.44, N = 3SE +/- 72.90, N = 3SE +/- 308.14, N = 3SE +/- 414.62, N = 3SE +/- 154.72, N = 3230970240702311056312009758403168251. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Xcompact3d Incompact3d

Input: input.i3d 193 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 193 Cells Per Directionc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.28, N = 3SE +/- 0.03, N = 3SE +/- 0.09, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 330.3125.8813.8313.7684.6613.95-pthread -ldl -lutil -lrt1. (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

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3100M200M300M400M500MSE +/- 392527.42, N = 3SE +/- 3333.33, N = 3SE +/- 3333.33, N = 3SE +/- 8819.17, N = 3SE +/- 92915.73, N = 3SE +/- 6666.67, N = 34600766671349266671627666671627566671295600001627533331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Liquid-DSP

Threads: 32 - Buffer Length: 256 - Filter Length: 512

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 32 - Buffer Length: 256 - Filter Length: 512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton360M120M180M240M300MSE +/- 193419.52, N = 3SE +/- 333.33, N = 3SE +/- 1000.00, N = 3SE +/- 577.35, N = 3SE +/- 120554.28, N = 3SE +/- 1855.92, N = 3274803333674863338141200081394000128250000813966671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Stress-NG

Test: Fused Multiply-Add

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Fused Multiply-Addc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton314M28M42M56M70MSE +/- 32747.05, N = 3SE +/- 3687.67, N = 3SE +/- 4431.60, N = 3SE +/- 10061.51, N = 3SE +/- 16948.60, N = 3SE +/- 4870.19, N = 330920910.9237732190.5463818458.6163723431.5510084119.2163762252.761. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Stress-NG

Test: Vector Shuffle

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Shufflec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton312K24K36K48K60KSE +/- 0.50, N = 3SE +/- 74.80, N = 3SE +/- 139.03, N = 3SE +/- 294.96, N = 3SE +/- 0.40, N = 3SE +/- 21.44, N = 322255.8435614.5154472.0754695.047162.5654143.401. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 32c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3500M1000M1500M2000M2500MSE +/- 218581.28, N = 3SE +/- 251661.15, N = 3SE +/- 284800.12, N = 3SE +/- 2915666.50, N = 3SE +/- 707515.21, N = 3SE +/- 435889.89, N = 3218486666715314000002271966667226683333364124333322705000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 64 - Buffer Length: 256 - Filter Length: 57c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3400M800M1200M1600M2000MSE +/- 1014889.16, N = 3SE +/- 11547.01, N = 3SE +/- 284800.12, N = 3SE +/- 88191.71, N = 3SE +/- 851162.60, N = 3SE +/- 152752.52, N = 317108000009782000001442366667144266666747789666714424000001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Stress-NG

Test: Matrix 3D Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix 3D Mathc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton32K4K6K8K10KSE +/- 1.96, N = 3SE +/- 1.40, N = 3SE +/- 9.35, N = 3SE +/- 19.16, N = 3SE +/- 9.17, N = 3SE +/- 6.38, N = 34571.965752.1710813.5910882.021841.5810403.93-laio -lbsd -lEGL -lGLESv2 -lmd1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Stress-NG

Test: Matrix Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Matrix Mathc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton380K160K240K320K400KSE +/- 167.77, N = 3SE +/- 8.13, N = 3SE +/- 38.76, N = 3SE +/- 28.60, N = 3SE +/- 6.88, N = 3SE +/- 53.44, N = 3147576.41284713.63368671.39369258.8955962.00368750.671. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Stress-NG

Test: Memory Copying

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Memory Copyingc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton34K8K12K16K20KSE +/- 0.46, N = 3SE +/- 1.12, N = 3SE +/- 4.65, N = 3SE +/- 1.36, N = 3SE +/- 1.93, N = 3SE +/- 3.80, N = 38080.4311324.7920478.6720475.963209.8320484.24-laio -lbsd -lEGL -lGLESv2 -lmd1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Liquid-DSP

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

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 32 - Buffer Length: 256 - Filter Length: 32c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3300M600M900M1200M1500MSE +/- 578311.72, N = 3SE +/- 456520.66, N = 3SE +/- 33333.33, N = 3SE +/- 57735.03, N = 3SE +/- 571382.34, N = 3SE +/- 233333.33, N = 311939666677654666671136133333113600000063348333311360666671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Liquid-DSP

Threads: 32 - Buffer Length: 256 - Filter Length: 57

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 1.6Threads: 32 - Buffer Length: 256 - Filter Length: 57c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3300M600M900M1200M1500MSE +/- 9533333.33, N = 3SE +/- 23094.01, N = 3SE +/- 168358.08, N = 3SE +/- 150111.07, N = 3SE +/- 846666.67, N = 3SE +/- 3333.33, N = 314442666674892700007213866677213800004295966677214933331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Stress-NG

Test: Vector Math

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.10Test: Vector Mathc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton350K100K150K200K250KSE +/- 100.78, N = 3SE +/- 37.96, N = 3SE +/- 20.95, N = 3SE +/- 27.00, N = 3SE +/- 9.74, N = 3SE +/- 47.94, N = 3221776.15147886.14217446.12217567.1038515.89217235.591. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz

Remhos

Test: Sample Remap Example

OpenBenchmarking.orgSeconds, Fewer Is BetterRemhos 1.0Test: Sample Remap Examplec6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton31530456075SE +/- 0.11, N = 3SE +/- 0.08, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.44, N = 3SE +/- 0.04, N = 322.1020.7414.1214.0868.9614.04-pthread1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi

Pennant

Test: sedovbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: sedovbigc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.036687, N = 3SE +/- 0.018218, N = 3SE +/- 0.011497, N = 3SE +/- 0.003721, N = 3SE +/- 0.050055, N = 3SE +/- 0.011347, N = 316.53050016.4805009.4222709.34095389.0049009.206490-pthread1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

Algebraic Multi-Grid Benchmark

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.2c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3400M800M1200M1600M2000MSE +/- 1055539.30, N = 3SE +/- 140169.34, N = 3SE +/- 192645.90, N = 3SE +/- 488508.39, N = 3SE +/- 394420.25, N = 3SE +/- 103191.30, N = 38369993001035586333176527766717659663334444561331646761667-pthread1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi

HeFFTe - Highly Efficient FFT for Exascale

Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton34080120160200SE +/- 0.08, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3SE +/- 0.13, N = 382.7681.94163.28163.5634.37162.96-pthread1. (CXX) g++ options: -O3

Monte Carlo Simulations of Ionised Nebulae

Input: Gas HII40

OpenBenchmarking.orgSeconds, Fewer Is BetterMonte Carlo Simulations of Ionised Nebulae 2.02.73.3Input: Gas HII40c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3612182430SE +/- 0.02, N = 3SE +/- 0.17, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 312.6720.7613.6613.5326.9213.58-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -cpp -Jsource/ -ffree-line-length-0 -lm -std=legacy -O2 -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lz

LULESH

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.3c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton36K12K18K24K30KSE +/- 90.11, N = 3SE +/- 38.55, N = 3SE +/- 11.81, N = 3SE +/- 12.73, N = 3SE +/- 5.42, N = 3SE +/- 27.09, N = 316708.2617557.4928708.6628736.235676.7828296.38-pthread1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi

Pennant

Test: leblancbig

OpenBenchmarking.orgHydro Cycle Time - Seconds, Fewer Is BetterPennant 1.0.1Test: leblancbigc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton31020304050SE +/- 0.013289, N = 3SE +/- 0.018924, N = 3SE +/- 0.005468, N = 3SE +/- 0.000467, N = 3SE +/- 0.025073, N = 3SE +/- 0.000869, N = 39.91756512.1768306.9613456.83999843.4588006.720537-pthread1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi

Xcompact3d Incompact3d

Input: input.i3d 129 Cells Per Direction

OpenBenchmarking.orgSeconds, Fewer Is BetterXcompact3d Incompact3d 2021-03-11Input: input.i3d 129 Cells Per Directionc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3510152025SE +/- 0.08686597, N = 15SE +/- 0.02560507, N = 3SE +/- 0.03233273, N = 3SE +/- 0.01738352, N = 3SE +/- 0.12655798, N = 3SE +/- 0.02702838, N = 37.019752885.637207353.144479993.1148982821.239273703.09871038-pthread -ldl -lutil -lrt1. (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

HeFFTe - Highly Efficient FFT for Exascale

Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3918273645SE +/- 0.17467, N = 3SE +/- 0.01033, N = 3SE +/- 0.02659, N = 3SE +/- 0.02971, N = 3SE +/- 0.01615, N = 3SE +/- 0.01031, N = 320.8719020.6279040.8283040.970809.3237940.89230-pthread1. (CXX) g++ options: -O3

NAS Parallel Benchmarks

Test / Class: CG.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: CG.Cc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton35K10K15K20K25KSE +/- 14.83, N = 3SE +/- 31.56, N = 3SE +/- 283.23, N = 3SE +/- 125.21, N = 3SE +/- 12.05, N = 3SE +/- 130.18, N = 320210.0013103.6221911.0222155.369661.9121988.99-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. c6a.16xlarge AMD Zen 3: Open MPI 4.1.23. egeo-07: Open MPI 4.1.0

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP CFD Solverc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3510152025SE +/- 0.002, N = 3SE +/- 0.016, N = 3SE +/- 0.021, N = 3SE +/- 0.027, N = 3SE +/- 0.208, N = 4SE +/- 0.011, N = 39.3426.0514.4424.42918.6304.375-O2 -lOpenCL-O2 -lOpenCL-O2 -lOpenCL-O2 -lOpenCL-m64 -lm -lcuda -lcudart -lcudadevrt -lcudart_static -lrt -lpthread -ldl-O2 -lOpenCL1. (CXX) g++ options:

HeFFTe - Highly Efficient FFT for Exascale

Test: c2c - Backend: FFTW - Precision: float - X Y Z: 256

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: float - X Y Z: 256c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.42, N = 6SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.10, N = 3SE +/- 0.01, N = 343.5941.9881.0181.1717.0081.44-pthread1. (CXX) g++ options: -O3

HeFFTe - Highly Efficient FFT for Exascale

Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton320406080100SE +/- 0.17, N = 3SE +/- 0.01, N = 3SE +/- 0.31, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 3SE +/- 0.02, N = 341.5940.1177.7778.1717.1278.50-pthread1. (CXX) g++ options: -O3

NAS Parallel Benchmarks

Test / Class: MG.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.Cc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton311K22K33K44K55KSE +/- 167.32, N = 3SE +/- 7.02, N = 3SE +/- 32.94, N = 3SE +/- 14.65, N = 3SE +/- 35.78, N = 3SE +/- 24.30, N = 345946.8125671.2949742.3049860.6819477.2350126.29-pthread -ldl -lutil -lrt1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. c6a.16xlarge AMD Zen 3: Open MPI 4.1.23. egeo-07: Open MPI 4.1.0

HeFFTe - Highly Efficient FFT for Exascale

Test: r2c - Backend: FFTW - Precision: float - X Y Z: 256

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: float - X Y Z: 256c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton34080120160200SE +/- 1.28, N = 3SE +/- 0.19, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 3SE +/- 0.04, N = 3SE +/- 0.27, N = 3102.6592.40162.01162.3632.43164.87-pthread1. (CXX) g++ options: -O3

LAMMPS Molecular Dynamics Simulator

Model: Rhodopsin Protein

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 23Jun2022Model: Rhodopsin Proteinc6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3918273645SE +/- 0.257, N = 12SE +/- 0.083, N = 3SE +/- 0.033, N = 3SE +/- 0.026, N = 3SE +/- 0.024, N = 3SE +/- 0.057, N = 319.56325.95037.41237.4827.17637.558-lm-pthread -lm1. (CXX) g++ options: -O3 -ldl

HeFFTe - Highly Efficient FFT for Exascale

Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton31326395265SE +/- 0.84547, N = 15SE +/- 0.08221, N = 3SE +/- 0.14885, N = 3SE +/- 0.32202, N = 3SE +/- 0.03519, N = 3SE +/- 0.28294, N = 348.9432032.7468055.1055055.103809.6229657.150301. (CXX) g++ options: -O3

HeFFTe - Highly Efficient FFT for Exascale

Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton370140210280350SE +/- 1.94, N = 3SE +/- 0.64, N = 3SE +/- 0.56, N = 3SE +/- 1.62, N = 3SE +/- 0.85, N = 15SE +/- 0.83, N = 3158.86209.50301.42300.4058.20306.54-pthread1. (CXX) g++ options: -O3

HeFFTe - Highly Efficient FFT for Exascale

Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton34080120160200SE +/- 1.25, N = 14SE +/- 0.35, N = 3SE +/- 0.47, N = 3SE +/- 0.20, N = 3SE +/- 0.04, N = 3SE +/- 0.27, N = 398.70135.36184.03184.1126.74186.36-pthread1. (CXX) g++ options: -O3

HeFFTe - Highly Efficient FFT for Exascale

Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128

OpenBenchmarking.orgGFLOP/s, More Is BetterHeFFTe - Highly Efficient FFT for Exascale 2.3Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128c6a.16xlarge AMD Zen 3c6g.16xlarge Graviton2c7g.16xlarge Graviton3c7gn.16xlarge Graviton3Eegeo-07m7g.16xlarge Graviton3306090120150SE +/- 1.46, N = 12SE +/- 0.61, N = 3SE +/- 0.47, N = 3SE +/- 0.04, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 386.3781.45133.51133.4223.02138.01-pthread1. (CXX) g++ options: -O3

Phoronix Test Suite v10.8.5