Benchmarks by Michael Larabel for a future article on Phoronix.com.
m7g.16xlarge Graviton3 Processor: ARMv8 Neoverse-V1 (64 Cores), Motherboard: Amazon EC2 m7g.16xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 256GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (aarch64), Compiler: GCC 11.3.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vPython Notes: Python 3.10.6Security Notes: 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 Changed Processor to ARMv8 Neoverse-N1 (64 Cores) .
Changed Motherboard to Amazon EC2 c6g.16xlarge (1.0 BIOS) .
Changed Memory to 128GB .
c7g.16xlarge Graviton3 Changed Processor to ARMv8 Neoverse-V1 (64 Cores) .
Changed Motherboard to Amazon EC2 c7g.16xlarge (1.0 BIOS) .
c7gn.16xlarge Graviton3E Changed Motherboard to Amazon EC2 c7gn.16xlarge (1.0 BIOS) .
c6a.16xlarge AMD Zen 3 Processor: AMD EPYC 7R13 (32 Cores / 64 Threads) , Motherboard: Amazon EC2 c6a.16xlarge (1.0 BIOS) , Chipset: Intel 440FX 82441FX PMC , Memory: 128GB, Disk: 322GB Amazon Elastic Block Store , Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (x86_64), Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vProcessor Notes: CPU Microcode: 0xa0011cfPython Notes: Python 3.10.12Security Notes: 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
Amazon AWS Graviton3E vs. Graviton 2/3 benchmarks Processor Motherboard Chipset Memory Disk Network OS Kernel Compiler File-System System Layer Vulkan m7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c6a.16xlarge AMD Zen 3 ARMv8 Neoverse-V1 (64 Cores) Amazon EC2 m7g.16xlarge (1.0 BIOS) Amazon Device 0200 256GB 215GB Amazon Elastic Block Store Amazon Elastic Ubuntu 22.04 5.19.0-1025-aws (aarch64) GCC 11.3.0 ext4 amazon ARMv8 Neoverse-N1 (64 Cores) Amazon EC2 c6g.16xlarge (1.0 BIOS) 128GB ARMv8 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 PMC 322GB Amazon Elastic Block Store 5.19.0-1025-aws (x86_64) 1.3.238 GCC 11.4.0 OpenBenchmarking.org Kernel Details - Transparent Huge Pages: madvise Compiler 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 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 Security 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 Processor Details - c6a.16xlarge AMD Zen 3: CPU Microcode: 0xa0011cf
m7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c6a.16xlarge AMD Zen 3 Result Overview Phoronix Test Suite 100% 142% 185% 227% 269% srsRAN Project ACES DGEMM Graph500 Liquid-DSP NAS Parallel Benchmarks OpenSSL Xcompact3d Incompact3d nekRS Algebraic Multi-Grid Benchmark Stress-NG LAMMPS Molecular Dynamics Simulator HeFFTe - Highly Efficient FFT for Exascale Pennant NWChem GROMACS LULESH nginx GPAW Monte Carlo Simulations of Ionised Nebulae BRL-CAD Kripke Remhos QMCPACK LeelaChessZero Timed Godot Game Engine Compilation Stockfish Rodinia Laghos 7-Zip Compression Coremark Timed Node.js Compilation Timed Gem5 Compilation
Amazon AWS Graviton3E vs. Graviton 2/3 benchmarks nwchem: C240 Buckyball lczero: BLAS brl-cad: VGR Performance Metric lczero: Eigen graph500: 26 graph500: 26 graph500: 26 graph500: 26 stockfish: Total Time lammps: 20k Atoms build-nodejs: Time To Compile qmcpack: FeCO6_b3lyp_gms qmcpack: FeCO6_b3lyp_gms build-gem5: Time To Compile openssl: SHA256 openssl: AES-128-GCM openssl: ChaCha20 openssl: ChaCha20-Poly1305 openssl: AES-256-GCM openssl: SHA512 mocassin: Dust 2D tau100.0 build-godot: Time To Compile stress-ng: CPU Cache nekrs: TurboPipe Periodic qmcpack: Li2_STO_ae nekrs: Kershaw npb: EP.D stress-ng: Wide Vector Math nginx: 1000 nginx: 500 npb: SP.C laghos: Sedov Blast Wave, ube_922_hex.mesh gpaw: Carbon Nanotube npb: LU.C coremark: CoreMark Size 666 - Iterations Per Second openssl: RSA4096 openssl: RSA4096 heffte: c2c - FFTW - double - 512 stress-ng: NUMA gromacs: MPI CPU - water_GMX50_bare rodinia: OpenMP LavaMD laghos: Triple Point Problem stress-ng: Vector Floating Point srsran: Downlink Processor Benchmark srsran: PUSCH Processor Benchmark, Throughput Total rodinia: OpenMP Streamcluster srsran: PUSCH Processor Benchmark, Throughput Thread qmcpack: simple-H2O heffte: r2c - FFTW - double - 512 kripke: heffte: c2c - FFTW - float - 512 liquid-dsp: 64 - 256 - 512 liquid-dsp: 32 - 256 - 512 stress-ng: Vector Shuffle stress-ng: Fused Multiply-Add stress-ng: Matrix Math liquid-dsp: 64 - 256 - 57 stress-ng: Matrix 3D Math stress-ng: Memory Copying liquid-dsp: 64 - 256 - 32 liquid-dsp: 32 - 256 - 32 liquid-dsp: 32 - 256 - 57 stress-ng: Vector Math compress-7zip: Decompression Rating compress-7zip: Compression Rating amg: mt-dgemm: Sustained Floating-Point Rate incompact3d: input.i3d 193 Cells Per Direction remhos: Sample Remap Example lulesh: heffte: r2c - FFTW - float - 512 mocassin: Gas HII40 pennant: sedovbig incompact3d: input.i3d 129 Cells Per Direction pennant: leblancbig npb: CG.C heffte: c2c - FFTW - double - 256 rodinia: OpenMP CFD Solver heffte: c2c - FFTW - float - 256 npb: MG.C heffte: r2c - FFTW - double - 256 lammps: Rhodopsin Protein heffte: r2c - FFTW - float - 256 heffte: c2c - FFTW - double - 128 heffte: c2c - FFTW - float - 128 heffte: r2c - FFTW - double - 128 heffte: r2c - FFTW - float - 128 m7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c6a.16xlarge AMD Zen 3 1940.2 1301 783777 1398 419754000 299497000 1227790000 1194320000 112119711 36.927 237.783 205.72 211.60 180.247 54212515580 332033171900 103226784517 74287460990 283333113630 32125448870 82.669 154.378 3892396.34 3976300000 112.61 3150680000 3738.98 1542834.94 255616.04 255768.44 17244.85 410.55 61.831 28341.68 1601880.342264 713859.5 10181.9 46.2504 3759.10 4.223 43.788 232.01 76102.55 318.5 5413.8 11.663 95.8 28.041 84.4739 339000400 88.0482 162753333 81396667 54143.40 63762252.76 368750.67 1442400000 10403.93 20484.24 2270500000 1136066667 721493333 217235.59 285540 316825 1646761667 24.362353 13.9454180 14.040 28296.378 162.956 13.575 9.206490 3.09871038 6.720537 21988.99 40.8923 4.375 81.4442 50126.29 78.5049 37.558 164.873 57.1503 186.356 138.014 306.540 2976.9 947 533020 891 284689000 209350000 874389000 860432000 86609284 25.171 287.814 297.94 302.19 225.305 42472798847 158436163857 67292541203 46717636807 129199593157 14393925490 145.374 218.276 1921785.20 2220190000 165.12 1760336667 2216.26 997272.65 158676.40 148964.69 9711.70 322.37 92.760 18741.90 1260642.177024 214040.9 2624.3 24.2658 2112.66 2.767 62.224 180.80 42850.82 197.2 3938.7 13.735 63.8 45.225 44.9297 220120233 42.8284 134926667 67486333 35614.51 37732190.54 284713.63 978200000 5752.17 11324.79 1531400000 765466667 489270000 147886.14 234202 240702 1035586333 20.417952 25.8825658 20.740 17557.485 81.9412 20.758 16.48050 5.63720735 12.17683 13103.62 20.6279 6.051 41.9816 25671.29 40.1104 25.950 92.3996 32.7468 135.358 81.4498 209.496 1962.7 1333 789066 1382 415758000 293826000 1206990000 1177710000 117316476 36.862 238.543 204.77 211.32 181.779 54216561263 332064349843 103275516997 74318842213 283373795737 32145914147 82.822 156.687 3844101.98 3978983333 112.64 3261853333 3664.54 1535336.57 255552.05 255145.52 17219.95 408.01 62.083 28375.71 1605948.674645 713945.9 10181.4 46.3706 3523.58 4.200 43.963 230.68 76178.46 319.7 5356.8 11.625 95.7 27.990 84.7451 354442733 88.1842 162766667 81412000 54472.07 63818458.61 368671.39 1442366667 10813.59 20478.67 2271966667 1136133333 721386667 217446.12 285633 311056 1765277667 24.140605 13.8326693 14.120 28708.656 163.276 13.659 9.422270 3.14447999 6.961345 21911.02 40.8283 4.442 81.0096 49742.30 77.7685 37.412 162.010 55.1055 184.026 133.514 301.418 1914 1392 744743 1444 411762000 296164000 1207760000 1175640000 117027121 36.838 238.636 204.25 188.28 182.471 54154218593 411130469943 114118119423 79969465487 351152465420 32126059040 82.974 155.951 3860335.38 4141440000 113.20 3302823333 3657.67 1530043.52 256585.83 253518.51 17163.11 423.11 56.440 28369.11 1611801.559265 713754.8 10183.3 46.5300 3525.17 4.820 44.044 236.22 76911.74 323.2 5431.2 10.690 97.4 27.999 85.0060 354234067 88.4551 162756667 81394000 54695.04 63723431.55 369258.89 1442666667 10882.02 20475.96 2266833333 1136000000 721380000 217567.10 285677 312009 1765966333 24.078529 13.7606726 14.082 28736.226 163.559 13.525 9.340953 3.11489828 6.839998 22155.36 40.9708 4.429 81.1671 49860.68 78.1658 37.482 162.361 55.1038 184.110 133.422 300.396 3440.4 1316 485038 1152 204550000 157688000 417777000 410571000 96905609 20.342 230.423 187.32 184.10 192.118 45857534777 151449269317 138389378753 92522999373 138457889450 15291283297 194.435 147.737 1447265.35 4337536667 123.95 4308810000 3061.42 1380146.63 163178.67 165847.75 34025.35 275.92 89.818 95221.40 1466587.036580 548396.5 8392.4 23.5212 552.68 3.965 64.179 227.40 96529.51 691.3 6479.1 8.396 215.9 26.867 42.4394 237087650 44.3176 460076667 274803333 22255.84 30920910.92 147576.41 1710800000 4571.96 8080.43 2184866667 1193966667 1444266667 221776.15 235787 230970 836999300 9.388050 30.3145288 22.104 16708.258 82.7584 12.669 16.53050 7.01975288 9.917565 20210.00 20.8719 9.342 43.5907 45946.81 41.5868 19.563 102.652 48.9432 98.7026 86.3730 158.858 OpenBenchmarking.org
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.org Seconds, Fewer Is Better NWChem 7.0.2 Input: C240 Buckyball m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 700 1400 2100 2800 3500 1940.2 1914.0 1962.7 2976.9 3440.4 -m64 1. (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
LeelaChessZero LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: BLAS m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300 600 900 1200 1500 SE +/- 4.67, N = 3 SE +/- 7.22, N = 3 SE +/- 3.53, N = 3 SE +/- 11.79, N = 3 SE +/- 13.29, N = 5 1301 1392 1333 947 1316 1. (CXX) g++ options: -flto -pthread
BRL-CAD BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.34 VGR Performance Metric m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 200K 400K 600K 800K 1000K 783777 744743 789066 533020 485038 -m64 1. (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
LeelaChessZero LeelaChessZero (lc0 / lczero) is a chess engine automated vian neural networks. This test profile can be used for OpenCL, CUDA + cuDNN, and BLAS (CPU-based) benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: Eigen m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300 600 900 1200 1500 SE +/- 8.74, N = 3 SE +/- 14.88, N = 3 SE +/- 15.65, N = 3 SE +/- 4.73, N = 3 SE +/- 7.37, N = 3 1398 1444 1382 891 1152 1. (CXX) g++ options: -flto -pthread
Graph500 This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org sssp max_TEPS, More Is Better Graph500 3.0 Scale: 26 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 90M 180M 270M 360M 450M 419754000 411762000 415758000 284689000 204550000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.org sssp median_TEPS, More Is Better Graph500 3.0 Scale: 26 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 60M 120M 180M 240M 300M 299497000 296164000 293826000 209350000 157688000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.org bfs max_TEPS, More Is Better Graph500 3.0 Scale: 26 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300M 600M 900M 1200M 1500M 1227790000 1207760000 1206990000 874389000 417777000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.org bfs median_TEPS, More Is Better Graph500 3.0 Scale: 26 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300M 600M 900M 1200M 1500M 1194320000 1175640000 1177710000 860432000 410571000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
Stockfish This is a test of Stockfish, an advanced open-source C++11 chess benchmark that can scale up to 512 CPU threads. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 15 Total Time m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 30M 60M 90M 120M 150M SE +/- 2854071.93, N = 15 SE +/- 1531345.46, N = 15 SE +/- 2998209.87, N = 12 SE +/- 2597495.37, N = 15 SE +/- 1430593.84, N = 15 112119711 117027121 117316476 86609284 96905609 -m64 -msse -msse3 -mpopcnt -mavx2 -msse4.1 -mssse3 -msse2 -mbmi2 1. (CXX) g++ options: -lgcov -lpthread -fno-exceptions -std=c++17 -fno-peel-loops -fno-tracer -pedantic -O3 -flto -flto=jobserver
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.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: 20k Atoms m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 36.93 36.84 36.86 25.17 20.34 -lm 1. (CXX) g++ options: -O3 -ldl
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.org Seconds, Fewer Is Better Timed Node.js Compilation 19.8.1 Time To Compile m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 60 120 180 240 300 SE +/- 0.33, N = 3 SE +/- 0.32, N = 3 SE +/- 0.20, N = 3 SE +/- 0.16, N = 3 SE +/- 0.40, N = 3 237.78 238.64 238.54 287.81 230.42
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.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.16 Input: FeCO6_b3lyp_gms m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 60 120 180 240 300 SE +/- 0.45, N = 3 SE +/- 0.21, N = 3 SE +/- 0.82, N = 3 SE +/- 1.75, N = 3 SE +/- 2.30, N = 3 205.72 204.25 204.77 297.94 187.32 -mcpu=native -mcpu=native -mcpu=native -mcpu=native -march=native 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl
OpenBenchmarking.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.16 Input: FeCO6_b3lyp_gms m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 70 140 210 280 350 SE +/- 0.22, N = 3 SE +/- 0.29, N = 3 SE +/- 0.19, N = 3 SE +/- 0.37, N = 3 SE +/- 1.03, N = 3 211.60 188.28 211.32 302.19 184.10 -mcpu=native -mcpu=native -mcpu=native -mcpu=native -march=native 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl
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.org Seconds, Fewer Is Better Timed Gem5 Compilation 21.2 Time To Compile m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50 100 150 200 250 SE +/- 0.13, N = 3 SE +/- 0.38, N = 3 SE +/- 0.26, N = 3 SE +/- 0.35, N = 3 SE +/- 0.26, N = 3 180.25 182.47 181.78 225.31 192.12
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA256 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 12000M 24000M 36000M 48000M 60000M SE +/- 18610524.10, N = 3 SE +/- 19542665.92, N = 3 SE +/- 16491036.11, N = 3 SE +/- 245440310.03, N = 3 SE +/- 26770675.21, N = 3 54212515580 54154218593 54216561263 42472798847 45857534777 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-128-GCM m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 90000M 180000M 270000M 360000M 450000M SE +/- 81289574.27, N = 3 SE +/- 11273100.69, N = 3 SE +/- 12264074.61, N = 3 SE +/- 9833681.11, N = 3 SE +/- 4227452.23, N = 3 332033171900 411130469943 332064349843 158436163857 151449269317 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 30000M 60000M 90000M 120000M 150000M SE +/- 1293723.80, N = 3 SE +/- 771581.87, N = 3 SE +/- 1725060.95, N = 3 SE +/- 35952887.59, N = 3 SE +/- 36376378.52, N = 3 103226784517 114118119423 103275516997 67292541203 138389378753 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: ChaCha20-Poly1305 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20000M 40000M 60000M 80000M 100000M SE +/- 1340503.89, N = 3 SE +/- 1769561.47, N = 3 SE +/- 1218886.42, N = 3 SE +/- 1132293.08, N = 3 SE +/- 232372675.93, N = 3 74287460990 79969465487 74318842213 46717636807 92522999373 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: AES-256-GCM m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 80000M 160000M 240000M 320000M 400000M SE +/- 6411836.47, N = 3 SE +/- 24279491.44, N = 3 SE +/- 33807617.40, N = 3 SE +/- 2312792.64, N = 3 SE +/- 41584947.90, N = 3 283333113630 351152465420 283373795737 129199593157 138457889450 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.1 Algorithm: SHA512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 7000M 14000M 21000M 28000M 35000M SE +/- 17714077.14, N = 3 SE +/- 16155877.53, N = 3 SE +/- 4573992.60, N = 3 SE +/- 9173912.49, N = 3 SE +/- 207279.55, N = 3 32125448870 32126059040 32145914147 14393925490 15291283297 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
Monte Carlo Simulations of Ionised Nebulae Mocassin is the Monte Carlo Simulations of Ionised Nebulae. MOCASSIN is a fully 3D or 2D photoionisation and dust radiative transfer code which employs a Monte Carlo approach to the transfer of radiation through media of arbitrary geometry and density distribution. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Monte Carlo Simulations of Ionised Nebulae 2.02.73.3 Input: Dust 2D tau100.0 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 40 80 120 160 200 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.00, N = 3 SE +/- 0.86, N = 3 SE +/- 1.84, N = 7 82.67 82.97 82.82 145.37 194.44 1. (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
Timed Godot Game Engine Compilation This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 4.0 Time To Compile m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50 100 150 200 250 SE +/- 0.32, N = 3 SE +/- 0.45, N = 3 SE +/- 0.63, N = 3 SE +/- 0.30, N = 3 SE +/- 0.12, N = 3 154.38 155.95 156.69 218.28 147.74
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: CPU Cache m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 800K 1600K 2400K 3200K 4000K SE +/- 57217.78, N = 15 SE +/- 40698.46, N = 15 SE +/- 59376.56, N = 15 SE +/- 21905.72, N = 15 SE +/- 30785.49, N = 12 3892396.34 3860335.38 3844101.98 1921785.20 1447265.35 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
nekRS nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming on smaller systems. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org flops/rank, More Is Better nekRS 23.0 Input: TurboPipe Periodic m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 900M 1800M 2700M 3600M 4500M SE +/- 1199180.28, N = 3 SE +/- 1394740.12, N = 3 SE +/- 169148.19, N = 3 SE +/- 144222.05, N = 3 SE +/- 12801180.07, N = 3 3976300000 4141440000 3978983333 2220190000 4337536667 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -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.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.16 Input: Li2_STO_ae m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 40 80 120 160 200 SE +/- 0.08, N = 3 SE +/- 0.31, N = 3 SE +/- 0.12, N = 3 SE +/- 1.13, N = 3 SE +/- 0.13, N = 3 112.61 113.20 112.64 165.12 123.95 -mcpu=native -mcpu=native -mcpu=native -mcpu=native -march=native 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl
nekRS nekRS is an open-source Navier Stokes solver based on the spectral element method. NekRS supports both CPU and GPU/accelerator support though this test profile is currently configured for CPU execution. NekRS is part of Nek5000 of the Mathematics and Computer Science MCS at Argonne National Laboratory. This nekRS benchmark is primarily relevant to large core count HPC servers and otherwise may be very time consuming on smaller systems. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org flops/rank, More Is Better nekRS 23.0 Input: Kershaw m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 900M 1800M 2700M 3600M 4500M SE +/- 1575066.14, N = 3 SE +/- 5414395.42, N = 3 SE +/- 2490845.46, N = 3 SE +/- 737119.02, N = 3 SE +/- 22342148.51, N = 3 3150680000 3302823333 3261853333 1760336667 4308810000 1. (CXX) g++ options: -fopenmp -O2 -march=native -mtune=native -ftree-vectorize -rdynamic -lmpi_cxx -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.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 800 1600 2400 3200 4000 SE +/- 1.69, N = 3 SE +/- 32.06, N = 15 SE +/- 34.07, N = 15 SE +/- 2.22, N = 3 SE +/- 4.77, N = 3 3738.98 3657.67 3664.54 2216.26 3061.42 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Wide Vector Math m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300K 600K 900K 1200K 1500K SE +/- 16116.93, N = 15 SE +/- 16444.95, N = 15 SE +/- 16521.46, N = 15 SE +/- 505.84, N = 3 SE +/- 2507.18, N = 3 1542834.94 1530043.52 1535336.57 997272.65 1380146.63 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
nginx This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 1000 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50K 100K 150K 200K 250K SE +/- 137.20, N = 3 SE +/- 402.16, N = 3 SE +/- 55.97, N = 3 SE +/- 185.79, N = 3 SE +/- 136.82, N = 3 255616.04 256585.83 255552.05 158676.40 163178.67 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.23.2 Connections: 500 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50K 100K 150K 200K 250K SE +/- 323.56, N = 3 SE +/- 317.05, N = 3 SE +/- 243.69, N = 3 SE +/- 90.87, N = 3 SE +/- 60.38, N = 3 255768.44 253518.51 255145.52 148964.69 165847.75 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2
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.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 7K 14K 21K 28K 35K SE +/- 10.19, N = 3 SE +/- 31.31, N = 3 SE +/- 7.21, N = 3 SE +/- 1.54, N = 3 SE +/- 20.85, N = 3 17244.85 17163.11 17219.95 9711.70 34025.35 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
Laghos Laghos (LAGrangian High-Order Solver) is a miniapp that solves the time-dependent Euler equations of compressible gas dynamics in a moving Lagrangian frame using unstructured high-order finite element spatial discretization and explicit high-order time-stepping. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Major Kernels Total Rate, More Is Better Laghos 3.1 Test: Sedov Blast Wave, ube_922_hex.mesh m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 90 180 270 360 450 SE +/- 0.42, N = 3 SE +/- 0.79, N = 3 SE +/- 0.89, N = 3 SE +/- 0.89, N = 3 SE +/- 0.48, N = 3 410.55 423.11 408.01 322.37 275.92 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
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.org Seconds, Fewer Is Better GPAW 23.6 Input: Carbon Nanotube m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20 40 60 80 100 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 61.83 56.44 62.08 92.76 89.82 1. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -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.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20K 40K 60K 80K 100K SE +/- 48.62, N = 3 SE +/- 43.73, N = 3 SE +/- 36.09, N = 3 SE +/- 26.12, N = 3 SE +/- 90.22, N = 3 28341.68 28369.11 28375.71 18741.90 95221.40 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
Coremark This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300K 600K 900K 1200K 1500K SE +/- 11449.37, N = 15 SE +/- 14869.41, N = 7 SE +/- 13274.76, N = 15 SE +/- 153.60, N = 3 SE +/- 6710.50, N = 3 1601880.34 1611801.56 1605948.67 1260642.18 1466587.04 1. (CC) gcc options: -O2 -lrt" -lrt
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 150K 300K 450K 600K 750K SE +/- 21.82, N = 3 SE +/- 198.10, N = 3 SE +/- 12.03, N = 3 SE +/- 88.30, N = 3 SE +/- 34.73, N = 3 713859.5 713754.8 713945.9 214040.9 548396.5 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.1 Algorithm: RSA4096 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 2K 4K 6K 8K 10K SE +/- 1.27, N = 3 SE +/- 0.84, N = 3 SE +/- 1.54, N = 3 SE +/- 1.71, N = 3 SE +/- 3.06, N = 3 10181.9 10183.3 10181.4 2624.3 8392.4 -m64 1. (CC) gcc options: -pthread -O3 -lssl -lcrypto -ldl
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double - X Y Z: 512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 11 22 33 44 55 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 46.25 46.53 46.37 24.27 23.52 1. (CXX) g++ options: -O3
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: NUMA m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 800 1600 2400 3200 4000 SE +/- 5.17, N = 3 SE +/- 7.31, N = 3 SE +/- 3.39, N = 3 SE +/- 1.53, N = 3 SE +/- 9.75, N = 15 3759.10 3525.17 3523.58 2112.66 552.68 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -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.org Ns Per Day, More Is Better GROMACS 2023 Implementation: MPI CPU - Input: water_GMX50_bare m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 1.0845 2.169 3.2535 4.338 5.4225 SE +/- 0.003, N = 3 SE +/- 0.003, N = 3 SE +/- 0.004, N = 3 SE +/- 0.002, N = 3 SE +/- 0.013, N = 3 4.223 4.820 4.200 2.767 3.965 1. (CXX) g++ options: -O3
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 14 28 42 56 70 SE +/- 0.15, N = 3 SE +/- 0.15, N = 3 SE +/- 0.11, N = 3 SE +/- 0.04, N = 3 SE +/- 0.53, N = 3 43.79 44.04 43.96 62.22 64.18 1. (CXX) g++ options: -O2 -lOpenCL
Laghos Laghos (LAGrangian High-Order Solver) is a miniapp that solves the time-dependent Euler equations of compressible gas dynamics in a moving Lagrangian frame using unstructured high-order finite element spatial discretization and explicit high-order time-stepping. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Major Kernels Total Rate, More Is Better Laghos 3.1 Test: Triple Point Problem m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50 100 150 200 250 SE +/- 0.28, N = 3 SE +/- 0.27, N = 3 SE +/- 0.16, N = 3 SE +/- 0.48, N = 3 SE +/- 1.06, N = 3 232.01 236.22 230.68 180.80 227.40 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Vector Floating Point m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20K 40K 60K 80K 100K SE +/- 190.19, N = 3 SE +/- 1.74, N = 3 SE +/- 71.97, N = 3 SE +/- 31.31, N = 3 SE +/- 864.23, N = 13 76102.55 76911.74 76178.46 42850.82 96529.51 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -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.org Mbps, More Is Better srsRAN Project 23.5 Test: Downlink Processor Benchmark m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 150 300 450 600 750 SE +/- 0.91, N = 3 SE +/- 0.06, N = 3 SE +/- 0.95, N = 3 SE +/- 0.25, N = 3 SE +/- 1.26, N = 3 318.5 323.2 319.7 197.2 691.3 -march=native -mfma 1. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -lgtest
OpenBenchmarking.org Mbps, More Is Better srsRAN Project 23.5 Test: PUSCH Processor Benchmark, Throughput Total m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 1400 2800 4200 5600 7000 SE +/- 4.08, N = 3 SE +/- 3.32, N = 3 SE +/- 1.80, N = 3 SE +/- 2.53, N = 3 SE +/- 21.76, N = 3 5413.8 5431.2 5356.8 3938.7 6479.1 -march=native -mfma 1. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -lgtest
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Streamcluster m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 4 8 12 16 20 SE +/- 0.138, N = 3 SE +/- 0.233, N = 12 SE +/- 0.099, N = 8 SE +/- 0.211, N = 15 SE +/- 0.101, N = 15 11.663 10.690 11.625 13.735 8.396 1. (CXX) g++ options: -O2 -lOpenCL
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.org Mbps, More Is Better srsRAN Project 23.5 Test: PUSCH Processor Benchmark, Throughput Thread m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50 100 150 200 250 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.55, N = 3 95.8 97.4 95.7 63.8 215.9 -march=native -mfma 1. (CXX) g++ options: -O3 -fno-trapping-math -fno-math-errno -lgtest
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.org Total Execution Time - Seconds, Fewer Is Better QMCPACK 3.16 Input: simple-H2O m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 10 20 30 40 50 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.24, N = 3 SE +/- 0.08, N = 3 28.04 28.00 27.99 45.23 26.87 -mcpu=native -mcpu=native -mcpu=native -mcpu=native -march=native 1. (CXX) g++ options: -fopenmp -foffload=disable -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -O3 -lm -ldl
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double - X Y Z: 512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 84.47 85.01 84.75 44.93 42.44 1. (CXX) g++ options: -O3
Kripke Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Throughput FoM, More Is Better Kripke 1.2.6 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 80M 160M 240M 320M 400M SE +/- 619419.33, N = 3 SE +/- 445212.18, N = 3 SE +/- 525406.56, N = 3 SE +/- 102787.75, N = 3 SE +/- 2932840.19, N = 4 339000400 354234067 354442733 220120233 237087650 1. (CXX) g++ options: -O3 -fopenmp -ldl
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: float - X Y Z: 512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.14, N = 3 88.05 88.46 88.18 42.83 44.32 1. (CXX) g++ options: -O3
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.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 100M 200M 300M 400M 500M SE +/- 6666.67, N = 3 SE +/- 8819.17, N = 3 SE +/- 3333.33, N = 3 SE +/- 3333.33, N = 3 SE +/- 392527.42, N = 3 162753333 162756667 162766667 134926667 460076667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 60M 120M 180M 240M 300M SE +/- 1855.92, N = 3 SE +/- 577.35, N = 3 SE +/- 1000.00, N = 3 SE +/- 333.33, N = 3 SE +/- 193419.52, N = 3 81396667 81394000 81412000 67486333 274803333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Vector Shuffle m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 12K 24K 36K 48K 60K SE +/- 21.44, N = 3 SE +/- 294.96, N = 3 SE +/- 139.03, N = 3 SE +/- 74.80, N = 3 SE +/- 0.50, N = 3 54143.40 54695.04 54472.07 35614.51 22255.84 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Fused Multiply-Add m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 14M 28M 42M 56M 70M SE +/- 4870.19, N = 3 SE +/- 10061.51, N = 3 SE +/- 4431.60, N = 3 SE +/- 3687.67, N = 3 SE +/- 32747.05, N = 3 63762252.76 63723431.55 63818458.61 37732190.54 30920910.92 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Matrix Math m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 80K 160K 240K 320K 400K SE +/- 53.44, N = 3 SE +/- 28.60, N = 3 SE +/- 38.76, N = 3 SE +/- 8.13, N = 3 SE +/- 167.77, N = 3 368750.67 369258.89 368671.39 284713.63 147576.41 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
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.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 57 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 400M 800M 1200M 1600M 2000M SE +/- 152752.52, N = 3 SE +/- 88191.71, N = 3 SE +/- 284800.12, N = 3 SE +/- 11547.01, N = 3 SE +/- 1014889.16, N = 3 1442400000 1442666667 1442366667 978200000 1710800000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Matrix 3D Math m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 2K 4K 6K 8K 10K SE +/- 6.38, N = 3 SE +/- 19.16, N = 3 SE +/- 9.35, N = 3 SE +/- 1.40, N = 3 SE +/- 1.96, N = 3 10403.93 10882.02 10813.59 5752.17 4571.96 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Memory Copying m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 4K 8K 12K 16K 20K SE +/- 3.80, N = 3 SE +/- 1.36, N = 3 SE +/- 4.65, N = 3 SE +/- 1.12, N = 3 SE +/- 0.46, N = 3 20484.24 20475.96 20478.67 11324.79 8080.43 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
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.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 64 - Buffer Length: 256 - Filter Length: 32 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 500M 1000M 1500M 2000M 2500M SE +/- 435889.89, N = 3 SE +/- 2915666.50, N = 3 SE +/- 284800.12, N = 3 SE +/- 251661.15, N = 3 SE +/- 218581.28, N = 3 2270500000 2266833333 2271966667 1531400000 2184866667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 32 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300M 600M 900M 1200M 1500M SE +/- 233333.33, N = 3 SE +/- 57735.03, N = 3 SE +/- 33333.33, N = 3 SE +/- 456520.66, N = 3 SE +/- 578311.72, N = 3 1136066667 1136000000 1136133333 765466667 1193966667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 1.6 Threads: 32 - Buffer Length: 256 - Filter Length: 57 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 300M 600M 900M 1200M 1500M SE +/- 3333.33, N = 3 SE +/- 150111.07, N = 3 SE +/- 168358.08, N = 3 SE +/- 23094.01, N = 3 SE +/- 9533333.33, N = 3 721493333 721380000 721386667 489270000 1444266667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Stress-NG Stress-NG is a Linux stress tool developed by Colin Ian King. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.15.10 Test: Vector Math m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 50K 100K 150K 200K 250K SE +/- 47.94, N = 3 SE +/- 27.00, N = 3 SE +/- 20.95, N = 3 SE +/- 37.96, N = 3 SE +/- 100.78, N = 3 217235.59 217567.10 217446.12 147886.14 221776.15 1. (CXX) g++ options: -lm -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lpthread -lrt -lsctp -lz
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.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 60K 120K 180K 240K 300K SE +/- 93.51, N = 3 SE +/- 54.90, N = 3 SE +/- 146.43, N = 3 SE +/- 15.43, N = 3 SE +/- 1190.65, N = 3 285540 285677 285633 234202 235787 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 70K 140K 210K 280K 350K SE +/- 154.72, N = 3 SE +/- 308.14, N = 3 SE +/- 72.90, N = 3 SE +/- 209.44, N = 3 SE +/- 670.46, N = 3 316825 312009 311056 240702 230970 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
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.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 400M 800M 1200M 1600M 2000M SE +/- 103191.30, N = 3 SE +/- 488508.39, N = 3 SE +/- 192645.90, N = 3 SE +/- 140169.34, N = 3 SE +/- 1055539.30, N = 3 1646761667 1765966333 1765277667 1035586333 836999300 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
ACES DGEMM This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 6 12 18 24 30 SE +/- 0.171001, N = 13 SE +/- 0.297525, N = 4 SE +/- 0.285590, N = 4 SE +/- 0.154503, N = 3 SE +/- 0.038051, N = 3 24.362353 24.078529 24.140605 20.417952 9.388050 1. (CC) gcc options: -O3 -march=native -fopenmp
Xcompact3d Incompact3d Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.28, N = 3 13.95 13.76 13.83 25.88 30.31 1. (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
Remhos Remhos (REMap High-Order Solver) is a miniapp that solves the pure advection equations that are used to perform monotonic and conservative discontinuous field interpolation (remap) as part of the Eulerian phase in Arbitrary Lagrangian Eulerian (ALE) simulations. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Remhos 1.0 Test: Sample Remap Example m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 5 10 15 20 25 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 14.04 14.08 14.12 20.74 22.10 1. (CXX) g++ options: -O3 -std=c++11 -lmfem -lHYPRE -lmetis -lrt -lmpi_cxx -lmpi
LULESH LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 6K 12K 18K 24K 30K SE +/- 27.09, N = 3 SE +/- 12.73, N = 3 SE +/- 11.81, N = 3 SE +/- 38.55, N = 3 SE +/- 90.11, N = 3 28296.38 28736.23 28708.66 17557.49 16708.26 1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: float - X Y Z: 512 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 40 80 120 160 200 SE +/- 0.13, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 162.96 163.56 163.28 81.94 82.76 1. (CXX) g++ options: -O3
Monte Carlo Simulations of Ionised Nebulae Mocassin is the Monte Carlo Simulations of Ionised Nebulae. MOCASSIN is a fully 3D or 2D photoionisation and dust radiative transfer code which employs a Monte Carlo approach to the transfer of radiation through media of arbitrary geometry and density distribution. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Monte Carlo Simulations of Ionised Nebulae 2.02.73.3 Input: Gas HII40 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.17, N = 3 SE +/- 0.02, N = 3 13.58 13.53 13.66 20.76 12.67 1. (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
Pennant Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 4 8 12 16 20 SE +/- 0.011347, N = 3 SE +/- 0.003721, N = 3 SE +/- 0.011497, N = 3 SE +/- 0.018218, N = 3 SE +/- 0.036687, N = 3 9.206490 9.340953 9.422270 16.480500 16.530500 1. (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.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 129 Cells Per Direction m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 2 4 6 8 10 SE +/- 0.02702838, N = 3 SE +/- 0.01738352, N = 3 SE +/- 0.03233273, N = 3 SE +/- 0.02560507, N = 3 SE +/- 0.08686597, N = 15 3.09871038 3.11489828 3.14447999 5.63720735 7.01975288 1. (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
Pennant Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: leblancbig m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 3 6 9 12 15 SE +/- 0.000869, N = 3 SE +/- 0.000467, N = 3 SE +/- 0.005468, N = 3 SE +/- 0.018924, N = 3 SE +/- 0.013289, N = 3 6.720537 6.839998 6.961345 12.176830 9.917565 1. (CXX) g++ options: -fopenmp -lmpi_cxx -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.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 5K 10K 15K 20K 25K SE +/- 130.18, N = 3 SE +/- 125.21, N = 3 SE +/- 283.23, N = 3 SE +/- 31.56, N = 3 SE +/- 14.83, N = 3 21988.99 22155.36 21911.02 13103.62 20210.00 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double - X Y Z: 256 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 9 18 27 36 45 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.17, N = 3 40.89 40.97 40.83 20.63 20.87 1. (CXX) g++ options: -O3
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP CFD Solver m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 3 6 9 12 15 SE +/- 0.011, N = 3 SE +/- 0.027, N = 3 SE +/- 0.021, N = 3 SE +/- 0.016, N = 3 SE +/- 0.002, N = 3 4.375 4.429 4.442 6.051 9.342 1. (CXX) g++ options: -O2 -lOpenCL
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: float - X Y Z: 256 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20 40 60 80 100 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.42, N = 6 81.44 81.17 81.01 41.98 43.59 1. (CXX) g++ options: -O3
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 11K 22K 33K 44K 55K SE +/- 24.30, N = 3 SE +/- 14.65, N = 3 SE +/- 32.94, N = 3 SE +/- 7.02, N = 3 SE +/- 167.32, N = 3 50126.29 49860.68 49742.30 25671.29 45946.81 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.2
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double - X Y Z: 256 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.31, N = 3 SE +/- 0.01, N = 3 SE +/- 0.17, N = 3 78.50 78.17 77.77 40.11 41.59 1. (CXX) g++ options: -O3
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.org ns/day, More Is Better LAMMPS Molecular Dynamics Simulator 23Jun2022 Model: Rhodopsin Protein m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 9 18 27 36 45 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.08, N = 3 SE +/- 0.26, N = 12 37.56 37.48 37.41 25.95 19.56 -lm 1. (CXX) g++ options: -O3 -ldl
HeFFTe - Highly Efficient FFT for Exascale HeFFTe is the Highly Efficient FFT for Exascale software developed as part of the Exascale Computing Project. This test profile uses HeFFTe's built-in speed benchmarks under a variety of configuration options and currently catering to CPU/processor testing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: float - X Y Z: 256 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 40 80 120 160 200 SE +/- 0.27, N = 3 SE +/- 0.04, N = 3 SE +/- 0.11, N = 3 SE +/- 0.19, N = 3 SE +/- 1.28, N = 3 164.87 162.36 162.01 92.40 102.65 1. (CXX) g++ options: -O3
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: double - X Y Z: 128 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 13 26 39 52 65 SE +/- 0.28, N = 3 SE +/- 0.32, N = 3 SE +/- 0.15, N = 3 SE +/- 0.08, N = 3 SE +/- 0.85, N = 15 57.15 55.10 55.11 32.75 48.94 1. (CXX) g++ options: -O3
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: c2c - Backend: FFTW - Precision: float - X Y Z: 128 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 40 80 120 160 200 SE +/- 0.27, N = 3 SE +/- 0.20, N = 3 SE +/- 0.47, N = 3 SE +/- 0.35, N = 3 SE +/- 1.25, N = 14 186.36 184.11 184.03 135.36 98.70 1. (CXX) g++ options: -O3
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: double - X Y Z: 128 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 30 60 90 120 150 SE +/- 0.12, N = 3 SE +/- 0.04, N = 3 SE +/- 0.47, N = 3 SE +/- 0.61, N = 3 SE +/- 1.46, N = 12 138.01 133.42 133.51 81.45 86.37 1. (CXX) g++ options: -O3
OpenBenchmarking.org GFLOP/s, More Is Better HeFFTe - Highly Efficient FFT for Exascale 2.3 Test: r2c - Backend: FFTW - Precision: float - X Y Z: 128 m7g.16xlarge Graviton3 c7gn.16xlarge Graviton3E c7g.16xlarge Graviton3 c6g.16xlarge Graviton2 c6a.16xlarge AMD Zen 3 70 140 210 280 350 SE +/- 0.83, N = 3 SE +/- 1.62, N = 3 SE +/- 0.56, N = 3 SE +/- 0.64, N = 3 SE +/- 1.94, N = 3 306.54 300.40 301.42 209.50 158.86 1. (CXX) g++ options: -O3
m7g.16xlarge Graviton3 Processor: ARMv8 Neoverse-V1 (64 Cores), Motherboard: Amazon EC2 m7g.16xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 256GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (aarch64), Compiler: GCC 11.3.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vPython Notes: Python 3.10.6Security Notes: 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
Testing initiated at 22 June 2023 16:24 by user ubuntu.
c6g.16xlarge Graviton2 Processor: ARMv8 Neoverse-N1 (64 Cores), Motherboard: Amazon EC2 c6g.16xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 128GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (aarch64), Compiler: GCC 11.3.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vPython Notes: Python 3.10.6Security Notes: 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
Testing initiated at 23 June 2023 01:32 by user ubuntu.
c7g.16xlarge Graviton3 Processor: ARMv8 Neoverse-V1 (64 Cores), Motherboard: Amazon EC2 c7g.16xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 128GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (aarch64), Compiler: GCC 11.3.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vPython Notes: Python 3.10.6Security Notes: 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
Testing initiated at 23 June 2023 10:31 by user ubuntu.
c7gn.16xlarge Graviton3E Processor: ARMv8 Neoverse-V1 (64 Cores), Motherboard: Amazon EC2 c7gn.16xlarge (1.0 BIOS), Chipset: Amazon Device 0200, Memory: 128GB, Disk: 215GB Amazon Elastic Block Store, Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (aarch64), Compiler: GCC 11.3.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vPython Notes: Python 3.10.6Security Notes: 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
Testing initiated at 10 July 2023 15:05 by user ubuntu.
c6a.16xlarge AMD Zen 3 Processor: AMD EPYC 7R13 (32 Cores / 64 Threads), Motherboard: Amazon EC2 c6a.16xlarge (1.0 BIOS), Chipset: Intel 440FX 82441FX PMC, Memory: 128GB, Disk: 322GB Amazon Elastic Block Store, Network: Amazon Elastic
OS: Ubuntu 22.04, Kernel: 5.19.0-1025-aws (x86_64), Vulkan: 1.3.238, Compiler: GCC 11.4.0, File-System: ext4, System Layer: amazon
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --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 -vProcessor Notes: CPU Microcode: 0xa0011cfPython Notes: Python 3.10.12Security Notes: 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
Testing initiated at 11 August 2023 14:59 by user ubuntu.