Tests for a future article on Phoronix.
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.
Result
OpenBenchmarking.org Seconds, Fewer Is Better GPAW 22.1 Input: Carbon Nanotube EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 SE +/- 0.18, N = 3 SE +/- 0.26, N = 13 SE +/- 0.21, N = 3 SE +/- 0.02, N = 3 SE +/- 0.35, N = 3 SE +/- 0.35, N = 4 SE +/- 0.19, N = 3 SE +/- 0.04, N = 3 40.29 39.22 24.65 23.50 35.83 32.55 23.15 21.01 1. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better GPAW 22.1 Input: Carbon Nanotube EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 8 16 24 32 40 Min: 40.03 / Avg: 40.29 / Max: 40.63 Min: 38.61 / Avg: 39.22 / Max: 42.27 Min: 24.39 / Avg: 24.65 / Max: 25.08 Min: 23.47 / Avg: 23.5 / Max: 23.54 Min: 35.3 / Avg: 35.83 / Max: 36.5 Min: 32.07 / Avg: 32.55 / Max: 33.6 Min: 22.94 / Avg: 23.15 / Max: 23.53 Min: 20.94 / Avg: 21.01 / Max: 21.09 1. (CC) gcc options: -shared -fwrapv -O2 -lxc -lblas -lmpi
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.
Result
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2022.1 Implementation: MPI CPU - Input: water_GMX50_bare EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 SE +/- 0.011, N = 3 SE +/- 0.020, N = 3 SE +/- 0.058, N = 3 SE +/- 0.022, N = 3 SE +/- 0.012, N = 3 SE +/- 0.010, N = 3 SE +/- 0.011, N = 3 SE +/- 1.013, N = 12 9.641 10.345 17.104 18.382 11.147 12.956 18.686 19.834 1. (CXX) g++ options: -O3 Result Confidence
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2022.1 Implementation: MPI CPU - Input: water_GMX50_bare EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 Min: 9.63 / Avg: 9.64 / Max: 9.66 Min: 10.31 / Avg: 10.35 / Max: 10.37 Min: 17.01 / Avg: 17.1 / Max: 17.21 Min: 18.34 / Avg: 18.38 / Max: 18.41 Min: 11.12 / Avg: 11.15 / Max: 11.17 Min: 12.94 / Avg: 12.96 / Max: 12.97 Min: 18.67 / Avg: 18.69 / Max: 18.71 Min: 8.76 / Avg: 19.83 / Max: 21.05 1. (CXX) g++ options: -O3
NAMD NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.0632 0.1264 0.1896 0.2528 0.316 SE +/- 0.00012, N = 4 SE +/- 0.00035, N = 4 SE +/- 0.00088, N = 3 SE +/- 0.00089, N = 5 SE +/- 0.00061, N = 3 SE +/- 0.00029, N = 4 SE +/- 0.00008, N = 3 SE +/- 0.00027, N = 3 0.28101 0.26207 0.17214 0.15564 0.24798 0.22157 0.12797 0.11211 Result Confidence
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.14 ATPase Simulation - 327,506 Atoms EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1 2 3 4 5 Min: 0.28 / Avg: 0.28 / Max: 0.28 Min: 0.26 / Avg: 0.26 / Max: 0.26 Min: 0.17 / Avg: 0.17 / Max: 0.17 Min: 0.15 / Avg: 0.16 / Max: 0.16 Min: 0.25 / Avg: 0.25 / Max: 0.25 Min: 0.22 / Avg: 0.22 / Max: 0.22 Min: 0.13 / Avg: 0.13 / Max: 0.13 Min: 0.11 / Avg: 0.11 / Max: 0.11
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 bfs median_TEPS, More Is Better Graph500 3.0 Scale: 26 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300M 600M 900M 1200M 1500M 799869000 797394000 1115180000 1238540000 887314000 863368000 1342540000 1479590000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.org bfs max_TEPS, More Is Better Graph500 3.0 Scale: 26 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300M 600M 900M 1200M 1500M 822265000 817075000 1192870000 1297530000 906613000 881729000 1437570000 1594850000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.org sssp median_TEPS, More Is Better Graph500 3.0 Scale: 26 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 130M 260M 390M 520M 650M 351607000 354905000 454926000 488130000 417648000 412741000 539862000 600906000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
OpenBenchmarking.org sssp max_TEPS, More Is Better Graph500 3.0 Scale: 26 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200M 400M 600M 800M 1000M 441522000 444031000 631819000 670721000 527751000 522933000 737889000 838159000 1. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 1.63, N = 12 SE +/- 0.03, N = 3 SE +/- 2.93, N = 10 SE +/- 0.37, N = 3 SE +/- 0.02, N = 3 SE +/- 0.45, N = 3 SE +/- 0.49, N = 3 SE +/- 0.13, N = 3 42.81 45.30 83.61 58.17 45.22 44.48 86.02 87.16 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi Result Confidence
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.1 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 30.08 / Avg: 42.81 / Max: 45.28 Min: 45.27 / Avg: 45.3 / Max: 45.36 Min: 57.32 / Avg: 83.61 / Max: 87.66 Min: 57.74 / Avg: 58.17 / Max: 58.9 Min: 45.19 / Avg: 45.22 / Max: 45.25 Min: 43.77 / Avg: 44.48 / Max: 45.31 Min: 85.06 / Avg: 86.02 / Max: 86.65 Min: 86.99 / Avg: 87.16 / Max: 87.41 1. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi
WRF WRF, the Weather Research and Forecasting Model, is a "next-generation mesoscale numerical weather prediction system designed for both atmospheric research and operational forecasting applications. It features two dynamical cores, a data assimilation system, and a software architecture supporting parallel computation and system extensibility." Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better WRF 4.2.2 Input: conus 2.5km EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K 8834.71 8828.50 4329.90 4503.42 7676.33 7370.39 4070.11 3940.08 1. (F9X) gfortran options: -O2 -ftree-vectorize -funroll-loops -ffree-form -fconvert=big-endian -frecord-marker=4 -fallow-invalid-boz -lesmf_time -lwrfio_nf -lnetcdff -lnetcdf -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz
RELION RELION - REgularised LIkelihood OptimisatioN - is a stand-alone computer program for Maximum A Posteriori refinement of (multiple) 3D reconstructions or 2D class averages in cryo-electron microscopy (cryo-EM). It is developed in the research group of Sjors Scheres at the MRC Laboratory of Molecular Biology. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better RELION 3.1.1 Test: Basic - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 60 120 180 240 300 SE +/- 2.72, N = 5 SE +/- 2.75, N = 5 SE +/- 0.15, N = 3 SE +/- 1.24, N = 12 SE +/- 2.71, N = 5 SE +/- 2.38, N = 6 SE +/- 0.19, N = 3 SE +/- 1.23, N = 12 256.41 250.95 131.60 130.65 247.09 245.03 127.52 127.04 1. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -lmpi_cxx -lmpi Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better RELION 3.1.1 Test: Basic - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50 100 150 200 250 Min: 253.42 / Avg: 256.41 / Max: 267.28 Min: 248.03 / Avg: 250.95 / Max: 261.97 Min: 131.34 / Avg: 131.59 / Max: 131.86 Min: 129.09 / Avg: 130.65 / Max: 144.26 Min: 244.21 / Avg: 247.09 / Max: 257.93 Min: 242.41 / Avg: 245.03 / Max: 256.91 Min: 127.3 / Avg: 127.52 / Max: 127.9 Min: 125.27 / Avg: 127.04 / Max: 140.5 1. (CXX) g++ options: -fopenmp -std=c++0x -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -lmpi_cxx -lmpi
Algebraic Multi-Grid Benchmark AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 900M 1800M 2700M 3600M 4500M SE +/- 2222006.08, N = 3 SE +/- 3446302.00, N = 3 SE +/- 13428026.69, N = 3 SE +/- 14635410.92, N = 3 SE +/- 1264541.59, N = 3 SE +/- 3283111.96, N = 3 SE +/- 3661935.20, N = 3 SE +/- 9793366.81, N = 3 2353380000 2355505667 4258717333 4279539333 2333430333 2341996333 4358309667 4414763333 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi Result Confidence
OpenBenchmarking.org Figure Of Merit, More Is Better Algebraic Multi-Grid Benchmark 1.2 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 800M 1600M 2400M 3200M 4000M Min: 2348944000 / Avg: 2353380000 / Max: 2355829000 Min: 2349661000 / Avg: 2355505666.67 / Max: 2361592000 Min: 4232661000 / Avg: 4258717333.33 / Max: 4277379000 Min: 4250591000 / Avg: 4279539333.33 / Max: 4297766000 Min: 2331796000 / Avg: 2333430333.33 / Max: 2335919000 Min: 2337896000 / Avg: 2341996333.33 / Max: 2348488000 Min: 4351993000 / Avg: 4358309666.67 / Max: 4364678000 Min: 4404357000 / Avg: 4414763333.33 / Max: 4434337000 1. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -lmpi
Xcompact3d Incompact3d Xcompact3d Incompact3d is a Fortran-MPI based, finite difference high-performance code for solving the incompressible Navier-Stokes equation and as many as you need scalar transport equations. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 129 Cells Per Direction EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.573 1.146 1.719 2.292 2.865 SE +/- 0.01662956, N = 9 SE +/- 0.01464622, N = 9 SE +/- 0.01020043, N = 10 SE +/- 0.01173139, N = 15 SE +/- 0.02850743, N = 15 SE +/- 0.01859828, N = 9 SE +/- 0.01070461, N = 10 SE +/- 0.01039810, N = 15 2.54647279 2.48146391 1.49829830 1.51555521 2.28903530 2.24137110 1.48172070 1.44810847 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 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 129 Cells Per Direction EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 2.46 / Avg: 2.55 / Max: 2.6 Min: 2.43 / Avg: 2.48 / Max: 2.56 Min: 1.46 / Avg: 1.5 / Max: 1.55 Min: 1.44 / Avg: 1.52 / Max: 1.6 Min: 2.06 / Avg: 2.29 / Max: 2.41 Min: 2.16 / Avg: 2.24 / Max: 2.34 Min: 1.42 / Avg: 1.48 / Max: 1.55 Min: 1.36 / Avg: 1.45 / Max: 1.53 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
Result
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 SE +/- 0.04408610, N = 5 SE +/- 0.02649159, N = 5 SE +/- 0.03372817, N = 7 SE +/- 0.02752118, N = 7 SE +/- 0.05206844, N = 5 SE +/- 0.02766243, N = 5 SE +/- 0.01966452, N = 7 SE +/- 0.01984368, N = 7 9.56905289 9.59882088 4.96696207 4.95660475 8.22620277 8.07779980 4.36289297 4.29938201 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 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: input.i3d 193 Cells Per Direction EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 9.47 / Avg: 9.57 / Max: 9.7 Min: 9.55 / Avg: 9.6 / Max: 9.67 Min: 4.84 / Avg: 4.97 / Max: 5.12 Min: 4.88 / Avg: 4.96 / Max: 5.04 Min: 8.09 / Avg: 8.23 / Max: 8.38 Min: 8.02 / Avg: 8.08 / Max: 8.17 Min: 4.28 / Avg: 4.36 / Max: 4.44 Min: 4.21 / Avg: 4.3 / Max: 4.38 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
Result
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 60 120 180 240 300 SE +/- 0.14, N = 3 SE +/- 0.71, N = 3 SE +/- 0.18, N = 3 SE +/- 0.14, N = 3 SE +/- 0.87, N = 3 SE +/- 0.89, N = 3 SE +/- 0.17, N = 3 SE +/- 0.23, N = 3 268.58 268.65 134.33 133.96 254.85 253.41 125.39 124.23 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 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Xcompact3d Incompact3d 2021-03-11 Input: X3D-benchmarking input.i3d EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50 100 150 200 250 Min: 268.42 / Avg: 268.58 / Max: 268.86 Min: 267.87 / Avg: 268.65 / Max: 270.06 Min: 134.15 / Avg: 134.33 / Max: 134.69 Min: 133.73 / Avg: 133.96 / Max: 134.22 Min: 253.9 / Avg: 254.85 / Max: 256.59 Min: 252.23 / Avg: 253.41 / Max: 255.15 Min: 125.2 / Avg: 125.39 / Max: 125.73 Min: 123.79 / Avg: 124.23 / Max: 124.6 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
LULESH LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 13K 26K 39K 52K 65K SE +/- 124.44, N = 4 SE +/- 122.69, N = 4 SE +/- 124.43, N = 4 SE +/- 99.91, N = 4 SE +/- 142.06, N = 4 SE +/- 141.39, N = 4 SE +/- 186.02, N = 4 SE +/- 227.17, N = 4 30166.92 30428.09 59398.79 60173.87 29508.59 30521.12 58017.07 59160.90 1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi Result Confidence
OpenBenchmarking.org z/s, More Is Better LULESH 2.0.3 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 10K 20K 30K 40K 50K Min: 29909.37 / Avg: 30166.92 / Max: 30498.81 Min: 30193.88 / Avg: 30428.09 / Max: 30653.55 Min: 59232.68 / Avg: 59398.79 / Max: 59762.39 Min: 59952.81 / Avg: 60173.87 / Max: 60413.95 Min: 29219.49 / Avg: 29508.59 / Max: 29898.49 Min: 30307.64 / Avg: 30521.12 / Max: 30927.36 Min: 57522.45 / Avg: 58017.07 / Max: 58368.87 Min: 58660.95 / Avg: 59160.9 / Max: 59712.29 1. (CXX) g++ options: -O3 -fopenmp -lm -lmpi_cxx -lmpi
Pennant Pennant is an application focused on hydrodynamics on general unstructured meshes in 2D. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: leblancbig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.9053 1.8106 2.7159 3.6212 4.5265 SE +/- 0.084956, N = 15 SE +/- 0.056585, N = 15 SE +/- 0.062550, N = 15 SE +/- 0.045794, N = 15 SE +/- 0.088305, N = 15 SE +/- 0.122551, N = 15 SE +/- 0.050393, N = 15 SE +/- 0.042439, N = 15 4.023470 3.919148 2.374382 2.375619 3.763021 3.382516 2.422264 2.146766 1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi Result Confidence
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: leblancbig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 3.67 / Avg: 4.02 / Max: 4.67 Min: 3.63 / Avg: 3.92 / Max: 4.42 Min: 2.06 / Avg: 2.37 / Max: 2.97 Min: 2.13 / Avg: 2.38 / Max: 2.71 Min: 3.3 / Avg: 3.76 / Max: 4.49 Min: 2.72 / Avg: 3.38 / Max: 4.32 Min: 2.11 / Avg: 2.42 / Max: 2.62 Min: 1.68 / Avg: 2.15 / Max: 2.25 1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi
Result
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 SE +/- 0.064993, N = 6 SE +/- 0.036194, N = 6 SE +/- 0.066125, N = 12 SE +/- 0.057833, N = 15 SE +/- 0.049036, N = 15 SE +/- 0.084204, N = 15 SE +/- 0.044883, N = 15 SE +/- 0.035923, N = 15 7.065542 6.936430 3.832093 3.816929 5.846230 5.211232 3.512603 3.053326 1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi Result Confidence
OpenBenchmarking.org Hydro Cycle Time - Seconds, Fewer Is Better Pennant 1.0.1 Test: sedovbig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 6.9 / Avg: 7.07 / Max: 7.3 Min: 6.86 / Avg: 6.94 / Max: 7.1 Min: 3.61 / Avg: 3.83 / Max: 4.42 Min: 3.51 / Avg: 3.82 / Max: 4.35 Min: 5.63 / Avg: 5.85 / Max: 6.22 Min: 4.82 / Avg: 5.21 / Max: 5.97 Min: 3.32 / Avg: 3.51 / Max: 3.9 Min: 2.85 / Avg: 3.05 / Max: 3.38 1. (CXX) g++ options: -fopenmp -lmpi_cxx -lmpi
miniFE MiniFE Finite Element is an application for unstructured implicit finite element codes. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org CG Mflops, More Is Better miniFE 2.2 Problem Size: Small EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 11K 22K 33K 44K 55K SE +/- 36.96, N = 5 SE +/- 35.08, N = 5 SE +/- 462.72, N = 15 SE +/- 352.42, N = 5 SE +/- 52.45, N = 5 SE +/- 108.17, N = 5 SE +/- 841.99, N = 15 SE +/- 1266.85, N = 15 51154.5 51282.3 43574.2 42074.2 51313.0 51284.7 45319.3 44206.6 1. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi Result Confidence
OpenBenchmarking.org CG Mflops, More Is Better miniFE 2.2 Problem Size: Small EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9K 18K 27K 36K 45K Min: 51054.8 / Avg: 51154.46 / Max: 51280.9 Min: 51209.3 / Avg: 51282.28 / Max: 51404.3 Min: 40240.1 / Avg: 43574.2 / Max: 46551.5 Min: 41607.6 / Avg: 42074.24 / Max: 43478.4 Min: 51215.5 / Avg: 51312.96 / Max: 51486.4 Min: 50975.2 / Avg: 51284.7 / Max: 51550.8 Min: 41157.2 / Avg: 45319.27 / Max: 50370.5 Min: 33101.3 / Avg: 44206.63 / Max: 50766.1 1. (CXX) g++ options: -O3 -fopenmp -lmpi_cxx -lmpi
ACES DGEMM This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.26, N = 6 SE +/- 0.25, N = 6 SE +/- 0.04, N = 8 SE +/- 0.06, N = 8 SE +/- 0.28, N = 7 SE +/- 0.17, N = 7 SE +/- 0.07, N = 9 SE +/- 0.08, N = 9 30.03 30.13 54.67 54.57 39.15 43.95 70.48 80.40 1. (CC) gcc options: -O3 -march=native -fopenmp Result Confidence
OpenBenchmarking.org GFLOP/s, More Is Better ACES DGEMM 1.0 Sustained Floating-Point Rate EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 15 30 45 60 75 Min: 28.82 / Avg: 30.03 / Max: 30.64 Min: 29.27 / Avg: 30.13 / Max: 30.98 Min: 54.49 / Avg: 54.67 / Max: 54.76 Min: 54.4 / Avg: 54.57 / Max: 54.9 Min: 38.42 / Avg: 39.15 / Max: 40.38 Min: 43.11 / Avg: 43.95 / Max: 44.42 Min: 70.04 / Avg: 70.48 / Max: 70.76 Min: 80.08 / Avg: 80.4 / Max: 80.78 1. (CC) gcc options: -O3 -march=native -fopenmp
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 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 1456.1 1441.5 1341.8 1309.8 1550.0 1391.0 1534.0 1355.8 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 -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -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.
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 120K 240K 360K 480K 600K SE +/- 455.33, N = 4 SE +/- 649.68, N = 4 SE +/- 3154.34, N = 6 SE +/- 4145.25, N = 6 SE +/- 1212.90, N = 4 SE +/- 1011.25, N = 5 SE +/- 2523.78, N = 6 SE +/- 3150.15, N = 6 235197.75 237735.42 463499.49 462177.87 267319.17 274024.34 500990.52 543554.45 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 90K 180K 270K 360K 450K Min: 234408.97 / Avg: 235197.75 / Max: 236099.24 Min: 236608 / Avg: 237735.42 / Max: 239535.63 Min: 451441.5 / Avg: 463499.49 / Max: 474719.49 Min: 444984.62 / Avg: 462177.87 / Max: 472625.12 Min: 265028.07 / Avg: 267319.17 / Max: 269568.84 Min: 271732.99 / Avg: 274024.34 / Max: 277051.61 Min: 493376.3 / Avg: 500990.52 / Max: 510934.82 Min: 533482.61 / Avg: 543554.45 / Max: 553168.85 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K SE +/- 131.05, N = 15 SE +/- 140.16, N = 15 SE +/- 108.53, N = 15 SE +/- 23.76, N = 12 SE +/- 86.35, N = 15 SE +/- 39.65, N = 15 SE +/- 112.63, N = 12 SE +/- 97.05, N = 15 7198.90 6940.40 12851.16 12962.47 8633.62 9551.94 16424.61 18773.11 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3K 6K 9K 12K 15K Min: 6450.89 / Avg: 7198.9 / Max: 7973.2 Min: 6441.7 / Avg: 6940.4 / Max: 7966.93 Min: 11720.55 / Avg: 12851.16 / Max: 13129.33 Min: 12854.11 / Avg: 12962.47 / Max: 13102.96 Min: 8304.72 / Avg: 8633.62 / Max: 9447.61 Min: 9431.7 / Avg: 9551.94 / Max: 10049.27 Min: 15551.37 / Avg: 16424.61 / Max: 16762.97 Min: 17448.4 / Avg: 18773.11 / Max: 19009.89 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5K 10K 15K 20K 25K SE +/- 165.09, N = 15 SE +/- 21.62, N = 4 SE +/- 571.15, N = 12 SE +/- 406.67, N = 15 SE +/- 56.69, N = 4 SE +/- 324.02, N = 12 SE +/- 249.56, N = 15 SE +/- 666.94, N = 12 8634.10 8887.57 16197.05 16347.89 11335.94 12299.79 20121.23 21722.65 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K Min: 6457.96 / Avg: 8634.1 / Max: 8926.65 Min: 8827.21 / Avg: 8887.57 / Max: 8923 Min: 12025.67 / Avg: 16197.05 / Max: 17807.02 Min: 12443.09 / Avg: 16347.89 / Max: 17750.34 Min: 11191.13 / Avg: 11335.94 / Max: 11451.46 Min: 9875.02 / Avg: 12299.79 / Max: 13116.29 Min: 18586.63 / Avg: 20121.23 / Max: 22325.57 Min: 17954.64 / Avg: 21722.65 / Max: 25144.84 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50K 100K 150K 200K 250K SE +/- 704.40, N = 8 SE +/- 565.95, N = 8 SE +/- 1508.85, N = 10 SE +/- 1649.35, N = 12 SE +/- 377.17, N = 8 SE +/- 986.09, N = 10 SE +/- 1534.56, N = 9 SE +/- 2780.73, N = 15 124561.13 127498.51 234088.25 234367.13 121799.77 125326.19 215057.58 222687.11 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40K 80K 120K 160K 200K Min: 121421.59 / Avg: 124561.13 / Max: 126802.64 Min: 125182.18 / Avg: 127498.51 / Max: 129865.62 Min: 222873.75 / Avg: 234088.25 / Max: 239534.97 Min: 227611.1 / Avg: 234367.13 / Max: 246966.17 Min: 120576.25 / Avg: 121799.77 / Max: 123852.21 Min: 119306.83 / Avg: 125326.19 / Max: 129242.95 Min: 209975.2 / Avg: 215057.58 / Max: 225878.46 Min: 199279.84 / Avg: 222687.11 / Max: 237740.52 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 110K 220K 330K 440K 550K SE +/- 702.52, N = 5 SE +/- 1412.50, N = 5 SE +/- 3622.62, N = 7 SE +/- 3515.49, N = 7 SE +/- 1225.90, N = 6 SE +/- 741.37, N = 6 SE +/- 3747.09, N = 7 SE +/- 4645.52, N = 7 254430.04 253395.11 471697.13 473377.17 274263.37 272077.94 490393.91 507823.13 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 90K 180K 270K 360K 450K Min: 252695.37 / Avg: 254430.04 / Max: 256269.6 Min: 248257.77 / Avg: 253395.11 / Max: 255823.7 Min: 457960.59 / Avg: 471697.13 / Max: 485078.19 Min: 458457.42 / Avg: 473377.17 / Max: 485079.54 Min: 271175.46 / Avg: 274263.37 / Max: 279773.24 Min: 269039.53 / Avg: 272077.94 / Max: 274139.5 Min: 472502.32 / Avg: 490393.91 / Max: 501177.83 Min: 491046.66 / Avg: 507823.13 / Max: 529808.26 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50K 100K 150K 200K 250K SE +/- 1837.39, N = 15 SE +/- 2022.23, N = 15 SE +/- 3991.38, N = 15 SE +/- 3716.81, N = 15 SE +/- 1433.38, N = 15 SE +/- 2245.97, N = 15 SE +/- 1703.57, N = 15 SE +/- 2791.46, N = 15 164817.94 167629.48 222846.94 214561.41 162758.97 168014.13 219454.01 238644.03 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40K 80K 120K 160K 200K Min: 147343.22 / Avg: 164817.94 / Max: 175088.74 Min: 155099.91 / Avg: 167629.48 / Max: 178196.4 Min: 198785.65 / Avg: 222846.94 / Max: 244849.44 Min: 197855.34 / Avg: 214561.41 / Max: 238551.1 Min: 153244.43 / Avg: 162758.97 / Max: 170206.52 Min: 154651.9 / Avg: 168014.13 / Max: 179048.24 Min: 210355.53 / Avg: 219454.01 / Max: 233725.92 Min: 224149.53 / Avg: 238644.03 / Max: 254369.77 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 60K 120K 180K 240K 300K SE +/- 341.83, N = 4 SE +/- 114.35, N = 4 SE +/- 525.61, N = 6 SE +/- 1312.27, N = 6 SE +/- 300.99, N = 4 SE +/- 276.27, N = 4 SE +/- 951.51, N = 6 SE +/- 1129.41, N = 6 119972.76 120312.32 241831.07 239623.83 125438.45 126790.36 257843.16 263182.48 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50K 100K 150K 200K 250K Min: 119029 / Avg: 119972.76 / Max: 120633.57 Min: 120038.45 / Avg: 120312.32 / Max: 120530.97 Min: 240679.66 / Avg: 241831.07 / Max: 243970.24 Min: 234160.72 / Avg: 239623.83 / Max: 243449.66 Min: 124952.09 / Avg: 125438.45 / Max: 126276.89 Min: 126092.46 / Avg: 126790.36 / Max: 127439.41 Min: 254333.01 / Avg: 257843.16 / Max: 260807.61 Min: 259352.16 / Avg: 263182.48 / Max: 266589.36 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 13.30, N = 5 SE +/- 8.06, N = 5 SE +/- 127.97, N = 15 SE +/- 146.90, N = 15 SE +/- 24.99, N = 5 SE +/- 18.47, N = 5 SE +/- 138.87, N = 15 SE +/- 190.79, N = 15 4878.23 4895.84 7690.07 8001.12 4888.52 4941.75 7694.93 8041.38 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1400 2800 4200 5600 7000 Min: 4828.71 / Avg: 4878.23 / Max: 4900.26 Min: 4875.6 / Avg: 4895.84 / Max: 4917.93 Min: 6838.28 / Avg: 7690.07 / Max: 8479.62 Min: 7234.06 / Avg: 8001.12 / Max: 8981.49 Min: 4834.4 / Avg: 4888.52 / Max: 4960.51 Min: 4895.9 / Avg: 4941.75 / Max: 4996.59 Min: 7196.96 / Avg: 7694.93 / Max: 8670.13 Min: 7104.22 / Avg: 8041.38 / Max: 9493.26 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50K 100K 150K 200K 250K SE +/- 1359.91, N = 15 SE +/- 1333.51, N = 15 SE +/- 990.19, N = 12 SE +/- 1222.97, N = 12 SE +/- 764.03, N = 15 SE +/- 1317.18, N = 15 SE +/- 1834.92, N = 15 SE +/- 2783.28, N = 15 125660.14 128432.87 233220.37 232995.13 120349.96 121727.31 197332.18 202432.69 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40K 80K 120K 160K 200K Min: 119606.09 / Avg: 125660.14 / Max: 136119.4 Min: 120209.2 / Avg: 128432.87 / Max: 135541.83 Min: 228934.33 / Avg: 233220.37 / Max: 241620.44 Min: 227070.36 / Avg: 232995.13 / Max: 241828.65 Min: 114792.34 / Avg: 120349.96 / Max: 126702.63 Min: 111490.34 / Avg: 121727.31 / Max: 131563.52 Min: 183412.3 / Avg: 197332.18 / Max: 209401.03 Min: 190499.39 / Avg: 202432.69 / Max: 228470.16 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 4.1.4
Result
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K SE +/- 289.00, N = 8 SE +/- 342.99, N = 8 SE +/- 964.25, N = 15 SE +/- 735.83, N = 15 SE +/- 546.79, N = 15 SE +/- 691.56, N = 15 SE +/- 1179.69, N = 15 SE +/- 875.77, N = 15 45631.34 46439.60 82493.40 81468.95 54251.02 55458.39 79107.08 82256.89 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 4.1.4 Result Confidence
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 14K 28K 42K 56K 70K Min: 44457.66 / Avg: 45631.34 / Max: 47009.76 Min: 44841.81 / Avg: 46439.6 / Max: 47857.7 Min: 76429.14 / Avg: 82493.4 / Max: 87689.83 Min: 76943.28 / Avg: 81468.95 / Max: 86375.38 Min: 49931.83 / Avg: 54251.02 / Max: 57549.97 Min: 50366.13 / Avg: 55458.39 / Max: 60141.66 Min: 72860.69 / Avg: 79107.08 / Max: 87207.29 Min: 75902.81 / Avg: 82256.89 / Max: 87774.98 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 4.1.4
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.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 8 16 24 32 40 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 4 SE +/- 0.03, N = 4 32.55 29.78 18.72 17.09 28.61 24.49 16.63 14.68 1. (CXX) g++ options: -O2 -lOpenCL Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 7 14 21 28 35 Min: 32.51 / Avg: 32.55 / Max: 32.58 Min: 29.72 / Avg: 29.78 / Max: 29.82 Min: 18.68 / Avg: 18.72 / Max: 18.74 Min: 17.02 / Avg: 17.09 / Max: 17.19 Min: 28.54 / Avg: 28.61 / Max: 28.71 Min: 24.41 / Avg: 24.49 / Max: 24.6 Min: 16.53 / Avg: 16.62 / Max: 16.71 Min: 14.63 / Avg: 14.68 / Max: 14.77 1. (CXX) g++ options: -O2 -lOpenCL
OpenVINO This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.25, N = 3 41.63 42.87 82.01 83.26 50.06 53.51 102.38 109.78 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 41.6 / Avg: 41.63 / Max: 41.66 Min: 42.83 / Avg: 42.87 / Max: 42.9 Min: 81.84 / Avg: 82.01 / Max: 82.16 Min: 83.13 / Avg: 83.26 / Max: 83.35 Min: 50.04 / Avg: 50.06 / Max: 50.08 Min: 53.47 / Avg: 53.51 / Max: 53.53 Min: 102.33 / Avg: 102.38 / Max: 102.47 Min: 109.33 / Avg: 109.78 / Max: 110.18 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 SE +/- 0.18, N = 3 SE +/- 0.26, N = 3 SE +/- 0.37, N = 3 SE +/- 0.12, N = 3 SE +/- 0.29, N = 3 SE +/- 0.45, N = 3 SE +/- 0.05, N = 3 SE +/- 1.05, N = 3 766.80 744.44 389.27 383.87 955.51 890.80 468.17 435.80 MIN: 733.6 / MAX: 808.13 MIN: 725.13 / MAX: 781.63 MIN: 380.46 / MAX: 438.65 MIN: 378.71 / MAX: 495.86 MIN: 860.92 / MAX: 1016.52 MIN: 402.28 / MAX: 953.71 MIN: 444.06 / MAX: 551.42 MIN: 388.31 / MAX: 645.46 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Face Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 Min: 766.44 / Avg: 766.8 / Max: 767 Min: 743.93 / Avg: 744.44 / Max: 744.72 Min: 388.66 / Avg: 389.27 / Max: 389.95 Min: 383.68 / Avg: 383.87 / Max: 384.08 Min: 955.09 / Avg: 955.51 / Max: 956.06 Min: 890.09 / Avg: 890.8 / Max: 891.62 Min: 468.11 / Avg: 468.17 / Max: 468.26 Min: 434.44 / Avg: 435.8 / Max: 437.87 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Face Detection FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50 100 150 200 250 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.15, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 SE +/- 0.12, N = 3 SE +/- 0.08, N = 3 80.62 83.27 156.27 158.29 96.87 103.49 194.03 212.43 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Face Detection FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 Min: 80.48 / Avg: 80.62 / Max: 80.73 Min: 83.1 / Avg: 83.27 / Max: 83.36 Min: 156.03 / Avg: 156.27 / Max: 156.55 Min: 158.24 / Avg: 158.29 / Max: 158.37 Min: 96.74 / Avg: 96.87 / Max: 96.97 Min: 103.34 / Avg: 103.49 / Max: 103.68 Min: 193.79 / Avg: 194.03 / Max: 194.18 Min: 212.29 / Avg: 212.43 / Max: 212.58 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40K 80K 120K 160K 200K SE +/- 69.14, N = 3 SE +/- 52.00, N = 3 SE +/- 1226.28, N = 3 SE +/- 1059.23, N = 3 SE +/- 105.34, N = 3 SE +/- 7.87, N = 3 SE +/- 625.00, N = 3 SE +/- 462.02, N = 3 97379.56 99729.06 161378.10 161750.11 112325.82 118385.58 144811.67 166260.62 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30K 60K 90K 120K 150K Min: 97247.58 / Avg: 97379.56 / Max: 97481.28 Min: 99674.93 / Avg: 99729.06 / Max: 99833.03 Min: 158960.14 / Avg: 161378.1 / Max: 162942.58 Min: 160490.05 / Avg: 161750.11 / Max: 163854.95 Min: 112115.64 / Avg: 112325.82 / Max: 112443.38 Min: 118371.42 / Avg: 118385.58 / Max: 118398.6 Min: 143768.01 / Avg: 144811.67 / Max: 145929.28 Min: 165479.5 / Avg: 166260.62 / Max: 167078.69 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.1463 0.2926 0.4389 0.5852 0.7315 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.52 0.51 0.51 0.51 0.65 0.62 0.54 0.52 MIN: 0.29 / MAX: 11.88 MIN: 0.29 / MAX: 16.07 MIN: 0.49 / MAX: 36.76 MIN: 0.49 / MAX: 33.4 MIN: 0.29 / MAX: 13.1 MIN: 0.29 / MAX: 13.83 MIN: 0.5 / MAX: 35 MIN: 0.5 / MAX: 33.28 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 0.52 / Avg: 0.52 / Max: 0.52 Min: 0.51 / Avg: 0.51 / Max: 0.51 Min: 0.51 / Avg: 0.51 / Max: 0.51 Min: 0.51 / Avg: 0.51 / Max: 0.51 Min: 0.65 / Avg: 0.65 / Max: 0.65 Min: 0.61 / Avg: 0.62 / Max: 0.62 Min: 0.54 / Avg: 0.54 / Max: 0.54 Min: 0.52 / Avg: 0.52 / Max: 0.52 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40K 80K 120K 160K 200K SE +/- 119.74, N = 3 SE +/- 114.30, N = 3 SE +/- 2309.74, N = 3 SE +/- 1524.49, N = 7 SE +/- 169.28, N = 3 SE +/- 151.20, N = 3 SE +/- 734.15, N = 3 SE +/- 731.54, N = 3 132051.79 133585.43 175383.10 171196.42 149560.47 159812.84 166269.03 173823.44 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30K 60K 90K 120K 150K Min: 131817 / Avg: 132051.79 / Max: 132210.03 Min: 133406.13 / Avg: 133585.43 / Max: 133797.88 Min: 172498.92 / Avg: 175383.1 / Max: 179950.23 Min: 166566.79 / Avg: 171196.42 / Max: 175720.49 Min: 149373.29 / Avg: 149560.47 / Max: 149898.38 Min: 159642.62 / Avg: 159812.84 / Max: 160114.41 Min: 164903.22 / Avg: 166269.03 / Max: 167418.66 Min: 173062.51 / Avg: 173823.44 / Max: 175286.13 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.1013 0.2026 0.3039 0.4052 0.5065 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 7 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.37 0.36 0.35 0.35 0.45 0.43 0.36 0.36 MIN: 0.21 / MAX: 11.31 MIN: 0.21 / MAX: 11.48 MIN: 0.33 / MAX: 35.74 MIN: 0.33 / MAX: 41.56 MIN: 0.21 / MAX: 22.82 MIN: 0.21 / MAX: 23.04 MIN: 0.34 / MAX: 35.18 MIN: 0.34 / MAX: 36.02 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1 2 3 4 5 Min: 0.36 / Avg: 0.37 / Max: 0.37 Min: 0.36 / Avg: 0.36 / Max: 0.36 Min: 0.35 / Avg: 0.35 / Max: 0.35 Min: 0.35 / Avg: 0.35 / Max: 0.35 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.42 / Avg: 0.43 / Max: 0.43 Min: 0.36 / Avg: 0.36 / Max: 0.36 Min: 0.35 / Avg: 0.36 / Max: 0.36 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 10 20 30 40 50 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 SE +/- 0.09, N = 3 SE +/- 0.03, N = 3 SE +/- 0.09, N = 3 SE +/- 0.43, N = 4 SE +/- 0.19, N = 3 23.07 23.90 36.22 37.16 27.74 29.74 38.04 41.74 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 Min: 22.96 / Avg: 23.07 / Max: 23.16 Min: 23.75 / Avg: 23.9 / Max: 24.02 Min: 35.92 / Avg: 36.22 / Max: 36.42 Min: 37.06 / Avg: 37.16 / Max: 37.35 Min: 27.69 / Avg: 27.74 / Max: 27.8 Min: 29.56 / Avg: 29.74 / Max: 29.84 Min: 37.28 / Avg: 38.04 / Max: 39.26 Min: 41.4 / Avg: 41.74 / Max: 42.05 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 400 800 1200 1600 2000 SE +/- 2.15, N = 3 SE +/- 3.90, N = 3 SE +/- 3.73, N = 3 SE +/- 1.97, N = 3 SE +/- 2.17, N = 3 SE +/- 4.15, N = 3 SE +/- 13.83, N = 4 SE +/- 5.11, N = 3 1373.83 1326.74 878.55 855.64 1714.98 1597.31 1253.98 1142.24 MIN: 791.41 / MAX: 1751.97 MIN: 804.84 / MAX: 1734.93 MIN: 733.22 / MAX: 1404.78 MIN: 729.63 / MAX: 1337.47 MIN: 907.6 / MAX: 2357.53 MIN: 894.43 / MAX: 2234.51 MIN: 831.98 / MAX: 2009.27 MIN: 789.57 / MAX: 1952.73 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 Min: 1370.69 / Avg: 1373.83 / Max: 1377.95 Min: 1320.87 / Avg: 1326.74 / Max: 1334.11 Min: 874.35 / Avg: 878.55 / Max: 885.98 Min: 851.73 / Avg: 855.64 / Max: 858 Min: 1710.64 / Avg: 1714.98 / Max: 1717.34 Min: 1592.31 / Avg: 1597.31 / Max: 1605.54 Min: 1214.86 / Avg: 1253.98 / Max: 1279.8 Min: 1133.53 / Avg: 1142.24 / Max: 1151.21 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 SE +/- 0.10, N = 3 SE +/- 0.04, N = 3 SE +/- 0.16, N = 3 SE +/- 0.36, N = 3 SE +/- 0.17, N = 3 SE +/- 0.11, N = 3 SE +/- 0.20, N = 3 SE +/- 0.22, N = 3 23.12 23.84 35.97 37.60 27.55 29.68 37.95 41.21 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 Min: 22.93 / Avg: 23.12 / Max: 23.28 Min: 23.77 / Avg: 23.84 / Max: 23.92 Min: 35.65 / Avg: 35.97 / Max: 36.13 Min: 36.97 / Avg: 37.6 / Max: 38.23 Min: 27.36 / Avg: 27.55 / Max: 27.88 Min: 29.52 / Avg: 29.68 / Max: 29.9 Min: 37.69 / Avg: 37.95 / Max: 38.35 Min: 40.97 / Avg: 41.21 / Max: 41.64 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 400 800 1200 1600 2000 SE +/- 5.54, N = 3 SE +/- 2.35, N = 3 SE +/- 4.19, N = 3 SE +/- 8.23, N = 3 SE +/- 9.11, N = 3 SE +/- 6.56, N = 3 SE +/- 7.07, N = 3 SE +/- 6.16, N = 3 1372.74 1330.39 884.80 846.31 1722.48 1598.26 1256.41 1157.04 MIN: 798.93 / MAX: 1782.56 MIN: 812.98 / MAX: 1776.36 MIN: 740.55 / MAX: 1368.43 MIN: 740.16 / MAX: 1340.88 MIN: 844.57 / MAX: 2377.74 MIN: 887.99 / MAX: 2186.68 MIN: 839.37 / MAX: 1915.41 MIN: 791.76 / MAX: 1935.94 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Detection FP32 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 Min: 1364.66 / Avg: 1372.74 / Max: 1383.34 Min: 1326.75 / Avg: 1330.39 / Max: 1334.79 Min: 880.2 / Avg: 884.8 / Max: 893.16 Min: 832.03 / Avg: 846.31 / Max: 860.55 Min: 1704.27 / Avg: 1722.48 / Max: 1732.13 Min: 1585.28 / Avg: 1598.26 / Max: 1606.33 Min: 1242.56 / Avg: 1256.41 / Max: 1265.79 Min: 1144.74 / Avg: 1157.04 / Max: 1163.92 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5K 10K 15K 20K 25K SE +/- 3.17, N = 3 SE +/- 2.84, N = 3 SE +/- 13.90, N = 3 SE +/- 13.29, N = 3 SE +/- 2.62, N = 3 SE +/- 2.97, N = 3 SE +/- 46.02, N = 3 SE +/- 17.72, N = 3 8228.70 8512.58 16203.41 16421.67 9842.93 10601.83 19812.09 21859.02 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K Min: 8222.81 / Avg: 8228.7 / Max: 8233.66 Min: 8508.49 / Avg: 8512.58 / Max: 8518.05 Min: 16177.98 / Avg: 16203.41 / Max: 16225.83 Min: 16397.26 / Avg: 16421.67 / Max: 16442.98 Min: 9838.31 / Avg: 9842.93 / Max: 9847.37 Min: 10595.89 / Avg: 10601.83 / Max: 10605.1 Min: 19757.3 / Avg: 19812.09 / Max: 19903.53 Min: 21823.81 / Avg: 21859.02 / Max: 21880.1 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 2.02, N = 3 SE +/- 1.00, N = 3 SE +/- 4.40, N = 3 SE +/- 0.65, N = 3 SE +/- 0.56, N = 3 SE +/- 2.77, N = 3 SE +/- 5.49, N = 3 SE +/- 7.19, N = 3 4165.02 4279.78 8090.58 8198.28 5004.25 5329.57 9979.66 10858.90 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K Min: 4162.93 / Avg: 4165.02 / Max: 4169.05 Min: 4277.9 / Avg: 4279.78 / Max: 4281.32 Min: 8083.99 / Avg: 8090.58 / Max: 8098.92 Min: 8196.99 / Avg: 8198.28 / Max: 8198.95 Min: 5003.68 / Avg: 5004.25 / Max: 5005.36 Min: 5325.59 / Avg: 5329.57 / Max: 5334.89 Min: 9970.73 / Avg: 9979.66 / Max: 9989.66 Min: 10846.86 / Avg: 10858.9 / Max: 10871.72 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 7.67 7.47 3.94 3.89 9.58 9.00 4.79 4.41 MIN: 4.04 / MAX: 17.18 MIN: 3.99 / MAX: 16.65 MIN: 3.84 / MAX: 19.45 MIN: 3.83 / MAX: 18.26 MIN: 4.88 / MAX: 27.41 MIN: 4.72 / MAX: 28.15 MIN: 3.96 / MAX: 28.62 MIN: 3.97 / MAX: 28.91 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Weld Porosity Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 7.66 / Avg: 7.67 / Max: 7.68 Min: 7.46 / Avg: 7.47 / Max: 7.47 Min: 3.93 / Avg: 3.94 / Max: 3.94 Min: 3.89 / Avg: 3.89 / Max: 3.89 Min: 9.58 / Avg: 9.58 / Max: 9.58 Min: 8.99 / Avg: 9 / Max: 9 Min: 4.79 / Avg: 4.79 / Max: 4.8 Min: 4.4 / Avg: 4.41 / Max: 4.41 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3K 6K 9K 12K 15K SE +/- 9.84, N = 3 SE +/- 16.12, N = 3 SE +/- 3.55, N = 3 SE +/- 11.42, N = 3 SE +/- 5.49, N = 3 SE +/- 7.76, N = 3 SE +/- 5.13, N = 3 SE +/- 3.81, N = 3 4978.42 5429.94 9148.55 9215.52 6003.73 6582.28 11178.65 12490.17 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16-INT8 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K Min: 4962.86 / Avg: 4978.42 / Max: 4996.64 Min: 5397.74 / Avg: 5429.94 / Max: 5447.34 Min: 9143.74 / Avg: 9148.55 / Max: 9155.48 Min: 9201.91 / Avg: 9215.52 / Max: 9238.22 Min: 5993.25 / Avg: 6003.73 / Max: 6011.83 Min: 6573.83 / Avg: 6582.28 / Max: 6597.78 Min: 11169.43 / Avg: 11178.65 / Max: 11187.15 Min: 12485.35 / Avg: 12490.17 / Max: 12497.68 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 3.40, N = 3 SE +/- 13.95, N = 3 SE +/- 1.39, N = 3 SE +/- 4.40, N = 3 SE +/- 10.61, N = 3 SE +/- 5.11, N = 3 SE +/- 1.33, N = 3 SE +/- 7.95, N = 3 2847.39 2880.48 6063.84 6310.01 3426.69 3414.62 7435.72 8084.61 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1400 2800 4200 5600 7000 Min: 2843.24 / Avg: 2847.39 / Max: 2854.12 Min: 2858.3 / Avg: 2880.48 / Max: 2906.22 Min: 6061.57 / Avg: 6063.84 / Max: 6066.37 Min: 6301.28 / Avg: 6310.01 / Max: 6315.29 Min: 3406.32 / Avg: 3426.69 / Max: 3442.03 Min: 3404.68 / Avg: 3414.62 / Max: 3421.64 Min: 7433.77 / Avg: 7435.72 / Max: 7438.26 Min: 8072.6 / Avg: 8084.61 / Max: 8099.65 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 11.23 11.10 5.27 5.06 13.99 14.04 6.44 5.93 MIN: 5.39 / MAX: 32.07 MIN: 5.35 / MAX: 31.25 MIN: 4.87 / MAX: 42.26 MIN: 4.86 / MAX: 45.19 MIN: 7.45 / MAX: 49.42 MIN: 5.88 / MAX: 52.59 MIN: 5.01 / MAX: 65.87 MIN: 4.9 / MAX: 68.82 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Vehicle Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 Min: 11.2 / Avg: 11.23 / Max: 11.24 Min: 11 / Avg: 11.1 / Max: 11.18 Min: 5.27 / Avg: 5.27 / Max: 5.27 Min: 5.06 / Avg: 5.06 / Max: 5.07 Min: 13.93 / Avg: 13.99 / Max: 14.08 Min: 14.02 / Avg: 14.04 / Max: 14.08 Min: 6.44 / Avg: 6.44 / Max: 6.44 Min: 5.92 / Avg: 5.93 / Max: 5.94 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 2.56, N = 3 SE +/- 12.97, N = 3 SE +/- 6.11, N = 3 SE +/- 5.54, N = 3 SE +/- 5.10, N = 3 SE +/- 8.43, N = 3 SE +/- 3.74, N = 3 SE +/- 0.76, N = 3 4762.61 4936.15 7395.02 7429.47 5763.31 6138.92 9010.78 9661.33 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K Min: 4757.69 / Avg: 4762.61 / Max: 4766.31 Min: 4912.48 / Avg: 4936.15 / Max: 4957.16 Min: 7384.74 / Avg: 7395.02 / Max: 7405.87 Min: 7418.81 / Avg: 7429.47 / Max: 7437.4 Min: 5754.58 / Avg: 5763.31 / Max: 5772.24 Min: 6122.18 / Avg: 6138.92 / Max: 6149 Min: 9005.57 / Avg: 9010.78 / Max: 9018.03 Min: 9660.08 / Avg: 9661.33 / Max: 9662.7 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 6.71 6.48 4.32 4.30 8.32 7.81 5.32 4.96 MIN: 4.88 / MAX: 19.92 MIN: 4.94 / MAX: 22.59 MIN: 4.15 / MAX: 27.82 MIN: 4.12 / MAX: 26.75 MIN: 5.2 / MAX: 32.51 MIN: 5.63 / MAX: 33.46 MIN: 4.36 / MAX: 47.85 MIN: 4.34 / MAX: 48.24 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Person Vehicle Bike Detection FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 6.7 / Avg: 6.71 / Max: 6.72 Min: 6.45 / Avg: 6.48 / Max: 6.51 Min: 4.31 / Avg: 4.32 / Max: 4.33 Min: 4.29 / Avg: 4.3 / Max: 4.31 Min: 8.3 / Avg: 8.32 / Max: 8.33 Min: 7.8 / Avg: 7.81 / Max: 7.83 Min: 5.31 / Avg: 5.32 / Max: 5.32 Min: 4.96 / Avg: 4.96 / Max: 4.96 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 SE +/- 0.82, N = 3 SE +/- 0.27, N = 3 SE +/- 0.64, N = 3 SE +/- 1.35, N = 3 SE +/- 0.51, N = 3 SE +/- 0.50, N = 3 SE +/- 0.69, N = 3 SE +/- 1.17, N = 3 449.76 470.91 792.53 808.16 547.98 590.08 973.45 1037.28 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org FPS, More Is Better OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 Min: 448.44 / Avg: 449.76 / Max: 451.26 Min: 470.63 / Avg: 470.91 / Max: 471.46 Min: 791.72 / Avg: 792.53 / Max: 793.79 Min: 805.47 / Avg: 808.16 / Max: 809.72 Min: 547.17 / Avg: 547.98 / Max: 548.91 Min: 589.55 / Avg: 590.08 / Max: 591.09 Min: 972.7 / Avg: 973.45 / Max: 974.83 Min: 1035.19 / Avg: 1037.28 / Max: 1039.25 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
Result
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.12, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 71.08 67.89 40.34 39.56 87.52 81.27 49.26 46.23 MIN: 37.67 / MAX: 102.95 MIN: 36.87 / MAX: 104.26 MIN: 34.61 / MAX: 149.47 MIN: 33.71 / MAX: 139.72 MIN: 46.08 / MAX: 141.07 MIN: 42.17 / MAX: 141.55 MIN: 38.21 / MAX: 191.1 MIN: 35.94 / MAX: 256.34 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OpenVINO 2022.2.dev Model: Machine Translation EN To DE FP16 - Device: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 70.87 / Avg: 71.08 / Max: 71.27 Min: 67.83 / Avg: 67.89 / Max: 67.94 Min: 40.27 / Avg: 40.34 / Max: 40.38 Min: 39.48 / Avg: 39.56 / Max: 39.69 Min: 87.36 / Avg: 87.52 / Max: 87.65 Min: 81.15 / Avg: 81.27 / Max: 81.33 Min: 49.19 / Avg: 49.26 / Max: 49.29 Min: 46.15 / Avg: 46.23 / Max: 46.32 1. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared
ASKAP ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Degridding EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K SE +/- 158.23, N = 3 SE +/- 270.04, N = 3 SE +/- 509.24, N = 3 SE +/- 307.17, N = 3 SE +/- 383.40, N = 3 SE +/- 223.70, N = 3 SE +/- 0.00, N = 3 SE +/- 721.42, N = 3 31647.3 31335.5 60853.9 62047.3 42366.7 46081.4 83966.6 88708.8 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Degridding EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 15K 30K 45K 60K 75K Min: 31330.8 / Avg: 31647.27 / Max: 31805.5 Min: 30870.1 / Avg: 31335.47 / Max: 31805.5 Min: 59976.1 / Avg: 60853.87 / Max: 61740.1 Min: 61740.1 / Avg: 62047.27 / Max: 62661.6 Min: 41983.3 / Avg: 42366.7 / Max: 43133.5 Min: 45634 / Avg: 46081.4 / Max: 46305.1 Min: 83966.6 / Avg: 83966.6 / Max: 83966.6 Min: 87465.2 / Avg: 88708.83 / Max: 89964.2 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
Result
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K SE +/- 0.00, N = 3 SE +/- 219.20, N = 3 SE +/- 696.55, N = 3 SE +/- 438.43, N = 3 SE +/- 626.96, N = 3 SE +/- 277.50, N = 3 SE +/- 460.77, N = 3 SE +/- 1025.23, N = 3 37485.1 37265.9 71171.8 74093.3 47485.2 51341.3 93531.7 98935.0 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp Result Confidence
OpenBenchmarking.org Mpix/sec, More Is Better ASKAP 1.0 Test: tConvolve MPI - Gridding EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K Min: 37485.1 / Avg: 37485.1 / Max: 37485.1 Min: 36827.5 / Avg: 37265.9 / Max: 37485.1 Min: 69972.2 / Avg: 71171.77 / Max: 72385 Min: 73654.9 / Avg: 74093.33 / Max: 74970.2 Min: 46305.1 / Avg: 47485.23 / Max: 48442.3 Min: 50786.3 / Avg: 51341.3 / Max: 51618.8 Min: 92610.2 / Avg: 93531.73 / Max: 93992.5 Min: 96884.5 / Avg: 98934.97 / Max: 99960.2 1. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
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.
Result
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: Eigen EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 105.96, N = 3 SE +/- 109.97, N = 3 SE +/- 22.88, N = 3 SE +/- 83.80, N = 3 SE +/- 107.55, N = 4 SE +/- 7.94, N = 3 SE +/- 125.11, N = 3 SE +/- 125.50, N = 4 8996 9607 9699 10188 9258 10155 9009 10387 1. (CXX) g++ options: -flto -pthread Result Confidence
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.28 Backend: Eigen EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K Min: 8791 / Avg: 8996 / Max: 9145 Min: 9393 / Avg: 9607 / Max: 9758 Min: 9653 / Avg: 9698.67 / Max: 9724 Min: 10077 / Avg: 10187.67 / Max: 10352 Min: 8945 / Avg: 9257.75 / Max: 9423 Min: 10143 / Avg: 10155 / Max: 10170 Min: 8775 / Avg: 9008.67 / Max: 9203 Min: 10245 / Avg: 10387.25 / Max: 10763 1. (CXX) g++ options: -flto -pthread
oneDNN This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of Intel oneAPI. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.151 0.302 0.453 0.604 0.755 SE +/- 0.000621, N = 7 SE +/- 0.001196, N = 7 SE +/- 0.001952, N = 7 SE +/- 0.002336, N = 7 SE +/- 0.002654, N = 7 SE +/- 0.001657, N = 7 SE +/- 0.004050, N = 7 SE +/- 0.004281, N = 7 0.595116 0.586275 0.439175 0.442635 0.670970 0.611600 0.525458 0.476241 MIN: 0.57 MIN: 0.56 MIN: 0.36 MIN: 0.37 MIN: 0.6 MIN: 0.58 MIN: 0.43 MIN: 0.41 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 0.59 / Avg: 0.6 / Max: 0.6 Min: 0.58 / Avg: 0.59 / Max: 0.59 Min: 0.43 / Avg: 0.44 / Max: 0.45 Min: 0.44 / Avg: 0.44 / Max: 0.45 Min: 0.66 / Avg: 0.67 / Max: 0.68 Min: 0.61 / Avg: 0.61 / Max: 0.62 Min: 0.5 / Avg: 0.53 / Max: 0.54 Min: 0.46 / Avg: 0.48 / Max: 0.49 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.0858 0.1716 0.2574 0.3432 0.429 SE +/- 0.000753, N = 7 SE +/- 0.001154, N = 7 SE +/- 0.002218, N = 15 SE +/- 0.001500, N = 7 SE +/- 0.001346, N = 7 SE +/- 0.001008, N = 7 SE +/- 0.000899, N = 7 SE +/- 0.002765, N = 15 0.280237 0.261029 0.289708 0.282014 0.285820 0.281104 0.381155 0.341806 MIN: 0.25 MIN: 0.24 MIN: 0.22 MIN: 0.21 MIN: 0.26 MIN: 0.25 MIN: 0.28 MIN: 0.27 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1 2 3 4 5 Min: 0.28 / Avg: 0.28 / Max: 0.28 Min: 0.26 / Avg: 0.26 / Max: 0.27 Min: 0.27 / Avg: 0.29 / Max: 0.3 Min: 0.28 / Avg: 0.28 / Max: 0.29 Min: 0.28 / Avg: 0.29 / Max: 0.29 Min: 0.28 / Avg: 0.28 / Max: 0.29 Min: 0.38 / Avg: 0.38 / Max: 0.38 Min: 0.32 / Avg: 0.34 / Max: 0.37 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.2572 0.5144 0.7716 1.0288 1.286 SE +/- 0.000646, N = 9 SE +/- 0.000782, N = 9 SE +/- 0.001947, N = 9 SE +/- 0.001619, N = 9 SE +/- 0.000926, N = 9 SE +/- 0.000988, N = 9 SE +/- 0.001925, N = 9 SE +/- 0.002404, N = 9 1.120080 1.115740 0.759754 0.745265 1.143180 1.089590 0.984734 0.938707 MIN: 1.11 MIN: 1.1 MIN: 0.73 MIN: 0.72 MIN: 1.04 MIN: 0.99 MIN: 0.9 MIN: 0.86 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.6 Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 1.12 / Avg: 1.12 / Max: 1.12 Min: 1.11 / Avg: 1.12 / Max: 1.12 Min: 0.75 / Avg: 0.76 / Max: 0.77 Min: 0.74 / Avg: 0.75 / Max: 0.75 Min: 1.14 / Avg: 1.14 / Max: 1.15 Min: 1.09 / Avg: 1.09 / Max: 1.09 Min: 0.98 / Avg: 0.98 / Max: 1 Min: 0.93 / Avg: 0.94 / Max: 0.96 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -std=c++11 -pie -ldl -lpthread
Timed LLVM Compilation This test times how long it takes to build the LLVM compiler stack. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 13.0 Build System: Ninja EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 SE +/- 0.56, N = 3 SE +/- 0.53, N = 3 SE +/- 0.21, N = 3 SE +/- 0.07, N = 3 SE +/- 0.11, N = 3 SE +/- 0.14, N = 3 SE +/- 0.18, N = 3 SE +/- 0.21, N = 3 117.76 115.77 83.34 81.68 105.19 97.62 74.65 71.33 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 13.0 Build System: Ninja EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 117 / Avg: 117.76 / Max: 118.84 Min: 114.71 / Avg: 115.77 / Max: 116.37 Min: 82.98 / Avg: 83.34 / Max: 83.7 Min: 81.6 / Avg: 81.68 / Max: 81.83 Min: 104.97 / Avg: 105.19 / Max: 105.33 Min: 97.45 / Avg: 97.62 / Max: 97.89 Min: 74.36 / Avg: 74.65 / Max: 74.98 Min: 71.04 / Avg: 71.33 / Max: 71.73
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 13.0 Build System: Unix Makefiles EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.36, N = 3 SE +/- 0.82, N = 3 SE +/- 0.41, N = 3 SE +/- 0.46, N = 3 SE +/- 0.19, N = 3 SE +/- 1.36, N = 3 SE +/- 0.11, N = 3 SE +/- 0.24, N = 3 180.48 178.75 159.74 159.93 171.12 167.35 158.69 157.88 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 13.0 Build System: Unix Makefiles EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 180.11 / Avg: 180.48 / Max: 181.21 Min: 177.61 / Avg: 178.75 / Max: 180.34 Min: 159.09 / Avg: 159.74 / Max: 160.49 Min: 159.01 / Avg: 159.93 / Max: 160.47 Min: 170.74 / Avg: 171.12 / Max: 171.33 Min: 165.75 / Avg: 167.35 / Max: 170.06 Min: 158.48 / Avg: 158.69 / Max: 158.81 Min: 157.51 / Avg: 157.88 / Max: 158.34
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.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Gem5 Compilation 21.2 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 SE +/- 1.10, N = 12 SE +/- 1.25, N = 3 SE +/- 1.11, N = 3 SE +/- 0.87, N = 3 SE +/- 2.00, N = 3 SE +/- 0.21, N = 3 SE +/- 0.76, N = 3 SE +/- 0.87, N = 3 150.77 148.97 135.99 134.83 146.31 141.44 137.08 135.26 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Gem5 Compilation 21.2 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 143.26 / Avg: 150.77 / Max: 154.1 Min: 146.99 / Avg: 148.97 / Max: 151.28 Min: 134.66 / Avg: 135.99 / Max: 138.19 Min: 133.17 / Avg: 134.83 / Max: 136.13 Min: 142.37 / Avg: 146.31 / Max: 148.86 Min: 141.08 / Avg: 141.44 / Max: 141.79 Min: 136.02 / Avg: 137.08 / Max: 138.56 Min: 133.6 / Avg: 135.26 / Max: 136.57
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.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 3.2.3 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 8 16 24 32 40 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 SE +/- 0.31, N = 3 SE +/- 0.38, N = 4 SE +/- 0.18, N = 3 SE +/- 0.07, N = 3 SE +/- 0.09, N = 3 SE +/- 0.38, N = 4 34.53 34.20 32.60 32.97 33.26 32.45 33.78 33.34 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Godot Game Engine Compilation 3.2.3 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 7 14 21 28 35 Min: 34.45 / Avg: 34.53 / Max: 34.64 Min: 33.94 / Avg: 34.2 / Max: 34.4 Min: 31.99 / Avg: 32.6 / Max: 33.02 Min: 31.9 / Avg: 32.97 / Max: 33.63 Min: 32.95 / Avg: 33.26 / Max: 33.58 Min: 32.36 / Avg: 32.45 / Max: 32.6 Min: 33.62 / Avg: 33.78 / Max: 33.95 Min: 32.41 / Avg: 33.34 / Max: 34.24
Timed Linux Kernel Compilation This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.18 Build: defconfig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 6 12 18 24 30 SE +/- 0.32, N = 3 SE +/- 0.24, N = 5 SE +/- 0.17, N = 7 SE +/- 0.18, N = 7 SE +/- 0.26, N = 5 SE +/- 0.23, N = 5 SE +/- 0.16, N = 9 SE +/- 0.15, N = 9 24.73 23.88 19.14 18.79 23.61 22.51 19.00 18.23 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.18 Build: defconfig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 6 12 18 24 30 Min: 24.34 / Avg: 24.73 / Max: 25.37 Min: 23.61 / Avg: 23.88 / Max: 24.84 Min: 18.89 / Avg: 19.14 / Max: 20.17 Min: 18.58 / Avg: 18.79 / Max: 19.85 Min: 23.28 / Avg: 23.61 / Max: 24.62 Min: 22.22 / Avg: 22.51 / Max: 23.41 Min: 18.74 / Avg: 19 / Max: 20.22 Min: 17.91 / Avg: 18.23 / Max: 19.33
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.18 Build: allmodconfig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.52, N = 3 SE +/- 0.42, N = 3 SE +/- 0.36, N = 3 SE +/- 0.39, N = 3 SE +/- 0.43, N = 3 SE +/- 0.49, N = 3 SE +/- 0.47, N = 3 SE +/- 0.08, N = 3 185.81 179.86 113.17 109.24 173.11 158.32 107.14 98.28 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 5.18 Build: allmodconfig EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 185.25 / Avg: 185.8 / Max: 186.84 Min: 179.34 / Avg: 179.86 / Max: 180.69 Min: 112.79 / Avg: 113.17 / Max: 113.89 Min: 108.83 / Avg: 109.24 / Max: 110.01 Min: 172.65 / Avg: 173.11 / Max: 173.97 Min: 157.71 / Avg: 158.32 / Max: 159.29 Min: 106.55 / Avg: 107.14 / Max: 108.07 Min: 98.13 / Avg: 98.28 / Max: 98.38
Timed Mesa Compilation This test profile times how long it takes to compile Mesa with Meson/Ninja. For minimizing build dependencies and avoid versioning conflicts, test this is just the core Mesa build without LLVM or the extra Gallium3D/Mesa drivers enabled. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Mesa Compilation 21.0 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.07, N = 3 SE +/- 0.19, N = 3 20.79 20.57 19.81 19.80 20.77 20.38 20.25 19.94 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Mesa Compilation 21.0 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 Min: 20.68 / Avg: 20.79 / Max: 20.9 Min: 20.51 / Avg: 20.57 / Max: 20.65 Min: 19.69 / Avg: 19.81 / Max: 19.98 Min: 19.72 / Avg: 19.8 / Max: 19.95 Min: 20.75 / Avg: 20.77 / Max: 20.81 Min: 20.35 / Avg: 20.38 / Max: 20.4 Min: 20.14 / Avg: 20.25 / Max: 20.38 Min: 19.74 / Avg: 19.94 / Max: 20.32
Timed MPlayer Compilation This test times how long it takes to build the MPlayer open-source media player program. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed MPlayer Compilation 1.5 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 SE +/- 0.014, N = 5 SE +/- 0.022, N = 5 SE +/- 0.024, N = 6 SE +/- 0.030, N = 6 SE +/- 0.006, N = 5 SE +/- 0.016, N = 5 SE +/- 0.029, N = 6 SE +/- 0.018, N = 6 9.162 8.947 7.795 7.676 8.765 8.414 7.697 7.493 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed MPlayer Compilation 1.5 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 9.12 / Avg: 9.16 / Max: 9.2 Min: 8.89 / Avg: 8.95 / Max: 9.01 Min: 7.73 / Avg: 7.8 / Max: 7.86 Min: 7.58 / Avg: 7.68 / Max: 7.79 Min: 8.75 / Avg: 8.77 / Max: 8.79 Min: 8.36 / Avg: 8.41 / Max: 8.46 Min: 7.6 / Avg: 7.7 / Max: 7.8 Min: 7.41 / Avg: 7.49 / Max: 7.54
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.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 18.8 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 SE +/- 0.32, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.11, N = 3 SE +/- 0.32, N = 3 SE +/- 0.18, N = 3 SE +/- 0.16, N = 3 SE +/- 0.37, N = 3 133.20 132.32 100.26 99.65 123.43 117.06 98.51 95.41 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Node.js Compilation 18.8 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 132.61 / Avg: 133.2 / Max: 133.69 Min: 132.28 / Avg: 132.32 / Max: 132.37 Min: 100.22 / Avg: 100.26 / Max: 100.29 Min: 99.48 / Avg: 99.65 / Max: 99.85 Min: 122.89 / Avg: 123.43 / Max: 123.98 Min: 116.71 / Avg: 117.06 / Max: 117.26 Min: 98.35 / Avg: 98.51 / Max: 98.83 Min: 94.81 / Avg: 95.41 / Max: 96.08
OpenBenchmarking.org Seconds, Fewer Is Better Timed CPython Compilation 3.10.6 Build Configuration: Released Build, PGO + LTO Optimized EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 60 120 180 240 300 251.68 251.21 253.34 253.27 254.59 254.61 255.75 255.70
Timed Wasmer Compilation This test times how long it takes to compile Wasmer. Wasmer is written in the Rust programming language and is a WebAssembly runtime implementation that supports WASI and EmScripten. This test profile builds Wasmer with the Cranelift and Singlepast compiler features enabled. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Timed Wasmer Compilation 2.3 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 SE +/- 0.11, N = 3 SE +/- 0.08, N = 3 SE +/- 0.04, N = 3 SE +/- 0.14, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.16, N = 3 40.10 40.18 40.72 40.86 40.45 40.41 41.09 41.44 1. (CC) gcc options: -m64 -ldl -lgcc_s -lutil -lrt -lpthread -lm -lc -pie -nodefaultlibs Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Timed Wasmer Compilation 2.3 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 Min: 39.93 / Avg: 40.1 / Max: 40.32 Min: 40.06 / Avg: 40.18 / Max: 40.34 Min: 40.65 / Avg: 40.72 / Max: 40.8 Min: 40.62 / Avg: 40.86 / Max: 41.11 Min: 40.25 / Avg: 40.45 / Max: 40.68 Min: 40.3 / Avg: 40.4 / Max: 40.58 Min: 41.03 / Avg: 41.09 / Max: 41.21 Min: 41.13 / Avg: 41.44 / Max: 41.65 1. (CC) gcc options: -m64 -ldl -lgcc_s -lutil -lrt -lpthread -lm -lc -pie -nodefaultlibs
Build2 This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.13 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 11 22 33 44 55 SE +/- 0.08, N = 3 SE +/- 0.10, N = 3 SE +/- 0.28, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 SE +/- 0.23, N = 3 SE +/- 0.23, N = 3 SE +/- 0.10, N = 3 50.86 50.71 48.32 48.49 50.60 49.53 48.78 48.73 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Build2 0.13 Time To Compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 10 20 30 40 50 Min: 50.72 / Avg: 50.86 / Max: 50.99 Min: 50.52 / Avg: 50.71 / Max: 50.85 Min: 47.76 / Avg: 48.32 / Max: 48.65 Min: 48.32 / Avg: 48.49 / Max: 48.71 Min: 50.53 / Avg: 50.6 / Max: 50.7 Min: 49.3 / Avg: 49.53 / Max: 49.99 Min: 48.5 / Avg: 48.78 / Max: 49.23 Min: 48.6 / Avg: 48.73 / Max: 48.92
DaCapo Benchmark This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 700 1400 2100 2800 3500 SE +/- 19.41, N = 8 SE +/- 12.71, N = 8 SE +/- 21.16, N = 8 SE +/- 18.24, N = 8 SE +/- 19.92, N = 8 SE +/- 16.67, N = 8 SE +/- 24.10, N = 8 SE +/- 22.93, N = 8 3205 3171 3302 3321 3240 3219 3380 3395 Result Confidence
OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 600 1200 1800 2400 3000 Min: 3138 / Avg: 3204.63 / Max: 3258 Min: 3134 / Avg: 3170.63 / Max: 3245 Min: 3231 / Avg: 3302.25 / Max: 3373 Min: 3209 / Avg: 3320.88 / Max: 3374 Min: 3188 / Avg: 3240.25 / Max: 3332 Min: 3184 / Avg: 3218.63 / Max: 3301 Min: 3287 / Avg: 3379.63 / Max: 3476 Min: 3314 / Avg: 3394.5 / Max: 3485
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.
Result
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200K 400K 600K 800K 1000K SE +/- 3953.49, N = 3 SE +/- 544.47, N = 3 SE +/- 6929.10, N = 3 SE +/- 3650.73, N = 3 SE +/- 1109.48, N = 3 SE +/- 903.59, N = 3 SE +/- 1019.34, N = 3 SE +/- 6383.59, N = 3 545728 551395 752608 762777 658923 700085 871440 890889 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC Result Confidence
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Compression Rating EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 150K 300K 450K 600K 750K Min: 537822 / Avg: 545728 / Max: 549789 Min: 550689 / Avg: 551395 / Max: 552466 Min: 738792 / Avg: 752608 / Max: 760452 Min: 755706 / Avg: 762777.33 / Max: 767888 Min: 656929 / Avg: 658923.33 / Max: 660763 Min: 698644 / Avg: 700085 / Max: 701750 Min: 869566 / Avg: 871440.33 / Max: 873072 Min: 884203 / Avg: 890888.67 / Max: 903651 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Result
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300K 600K 900K 1200K 1500K SE +/- 388.18, N = 3 SE +/- 950.26, N = 3 SE +/- 2445.95, N = 3 SE +/- 14137.94, N = 3 SE +/- 105.81, N = 3 SE +/- 2286.04, N = 3 SE +/- 14710.35, N = 3 SE +/- 6319.46, N = 3 523531 524476 943273 947566 652970 741536 1166359 1332056 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC Result Confidence
OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 22.01 Test: Decompression Rating EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200K 400K 600K 800K 1000K Min: 522879 / Avg: 523531 / Max: 524222 Min: 522810 / Avg: 524476.33 / Max: 526101 Min: 938381 / Avg: 943272.67 / Max: 945760 Min: 921112 / Avg: 947565.67 / Max: 969441 Min: 652781 / Avg: 652969.67 / Max: 653147 Min: 738348 / Avg: 741535.67 / Max: 745968 Min: 1137161 / Avg: 1166359 / Max: 1184087 Min: 1321179 / Avg: 1332056.33 / Max: 1343069 1. (CXX) g++ options: -lpthread -ldl -O2 -fPIC
Zstd Compression This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Compression Speed EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.23, N = 3 SE +/- 0.31, N = 3 SE +/- 1.46, N = 3 SE +/- 1.36, N = 15 SE +/- 0.23, N = 3 SE +/- 1.03, N = 3 SE +/- 0.35, N = 3 SE +/- 0.61, N = 3 103.3 102.1 103.3 102.9 106.1 107.3 108.4 107.8 1. (CC) gcc options: -O3 -pthread -lz -llzma Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19 - Compression Speed EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 102.9 / Avg: 103.3 / Max: 103.7 Min: 101.5 / Avg: 102.1 / Max: 102.5 Min: 100.4 / Avg: 103.3 / Max: 105.1 Min: 92.8 / Avg: 102.92 / Max: 113.3 Min: 105.6 / Avg: 106.07 / Max: 106.3 Min: 105.3 / Avg: 107.33 / Max: 108.6 Min: 107.8 / Avg: 108.4 / Max: 109 Min: 106.8 / Avg: 107.83 / Max: 108.9 1. (CC) gcc options: -O3 -pthread -lz -llzma
Result
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 800 1600 2400 3200 4000 SE +/- 40.76, N = 3 SE +/- 5.81, N = 3 SE +/- 18.13, N = 12 SE +/- 30.54, N = 12 SE +/- 1.24, N = 3 SE +/- 39.99, N = 3 SE +/- 11.04, N = 15 SE +/- 21.16, N = 12 3926.0 3837.0 3870.9 3843.5 3798.1 3852.8 3797.6 3779.6 1. (CC) gcc options: -O3 -pthread -lz -llzma Result Confidence
OpenBenchmarking.org MB/s, More Is Better Zstd Compression 1.5.0 Compression Level: 19, Long Mode - Decompression Speed EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 700 1400 2100 2800 3500 Min: 3846 / Avg: 3925.97 / Max: 3979.7 Min: 3830.3 / Avg: 3837.03 / Max: 3848.6 Min: 3813.9 / Avg: 3870.85 / Max: 3981.6 Min: 3565.5 / Avg: 3843.51 / Max: 3971.3 Min: 3796.4 / Avg: 3798.07 / Max: 3800.5 Min: 3773 / Avg: 3852.77 / Max: 3897.7 Min: 3748.6 / Avg: 3797.64 / Max: 3906.8 Min: 3609.2 / Avg: 3779.58 / Max: 3905.2 1. (CC) gcc options: -O3 -pthread -lz -llzma
Blender Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: BMW27 - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 SE +/- 0.15, N = 4 SE +/- 0.04, N = 4 SE +/- 0.03, N = 5 SE +/- 0.02, N = 6 18.39 17.45 10.01 9.26 15.63 13.59 8.43 7.50 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: BMW27 - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 Min: 18.33 / Avg: 18.39 / Max: 18.49 Min: 17.34 / Avg: 17.45 / Max: 17.6 Min: 9.97 / Avg: 10.01 / Max: 10.06 Min: 9.23 / Avg: 9.26 / Max: 9.32 Min: 15.45 / Avg: 15.63 / Max: 16.07 Min: 13.5 / Avg: 13.59 / Max: 13.67 Min: 8.36 / Avg: 8.43 / Max: 8.53 Min: 7.45 / Avg: 7.5 / Max: 7.58
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Classroom - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 10 20 30 40 50 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 46.00 43.27 24.43 22.30 38.72 33.39 20.12 17.66 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Classroom - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9 18 27 36 45 Min: 45.9 / Avg: 46 / Max: 46.06 Min: 43.25 / Avg: 43.27 / Max: 43.28 Min: 24.4 / Avg: 24.43 / Max: 24.46 Min: 22.25 / Avg: 22.3 / Max: 22.36 Min: 38.61 / Avg: 38.72 / Max: 38.78 Min: 33.34 / Avg: 33.39 / Max: 33.49 Min: 20.09 / Avg: 20.12 / Max: 20.14 Min: 17.61 / Avg: 17.66 / Max: 17.69
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Fishy Cat - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 6 12 18 24 30 SE +/- 0.07, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 4 SE +/- 0.06, N = 4 SE +/- 0.13, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 4 SE +/- 0.06, N = 5 24.00 22.59 13.49 12.32 20.47 18.05 11.50 10.54 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Fishy Cat - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 6 12 18 24 30 Min: 23.92 / Avg: 24 / Max: 24.15 Min: 22.5 / Avg: 22.59 / Max: 22.64 Min: 13.33 / Avg: 13.49 / Max: 13.58 Min: 12.21 / Avg: 12.32 / Max: 12.44 Min: 20.25 / Avg: 20.47 / Max: 20.69 Min: 18.02 / Avg: 18.05 / Max: 18.08 Min: 11.37 / Avg: 11.5 / Max: 11.6 Min: 10.41 / Avg: 10.54 / Max: 10.74
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Pabellon Barcelona - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 13 26 39 52 65 SE +/- 0.08, N = 3 SE +/- 0.28, N = 3 SE +/- 0.10, N = 3 SE +/- 0.33, N = 3 SE +/- 0.07, N = 3 SE +/- 0.12, N = 3 SE +/- 0.02, N = 3 SE +/- 0.20, N = 3 58.18 54.76 31.11 28.57 48.89 43.03 25.71 23.48 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Pabellon Barcelona - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 12 24 36 48 60 Min: 58.08 / Avg: 58.18 / Max: 58.34 Min: 54.43 / Avg: 54.76 / Max: 55.32 Min: 30.93 / Avg: 31.11 / Max: 31.27 Min: 28.23 / Avg: 28.57 / Max: 29.22 Min: 48.81 / Avg: 48.89 / Max: 49.03 Min: 42.83 / Avg: 43.03 / Max: 43.23 Min: 25.67 / Avg: 25.71 / Max: 25.74 Min: 23.26 / Avg: 23.48 / Max: 23.88
Result
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Barbershop - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.07, N = 3 SE +/- 0.18, N = 3 SE +/- 0.20, N = 3 SE +/- 0.11, N = 3 SE +/- 0.09, N = 3 SE +/- 0.38, N = 3 SE +/- 0.37, N = 3 SE +/- 0.72, N = 3 172.18 164.66 97.97 89.46 143.31 127.11 81.56 73.39 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Blender 3.3 Blend File: Barbershop - Compute: CPU-Only EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 172.05 / Avg: 172.18 / Max: 172.25 Min: 164.4 / Avg: 164.66 / Max: 165.02 Min: 97.71 / Avg: 97.97 / Max: 98.36 Min: 89.3 / Avg: 89.46 / Max: 89.68 Min: 143.14 / Avg: 143.31 / Max: 143.42 Min: 126.72 / Avg: 127.11 / Max: 127.86 Min: 80.87 / Avg: 81.56 / Max: 82.12 Min: 72.45 / Avg: 73.39 / Max: 74.8
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 SE +/- 0.016, N = 5 SE +/- 0.021, N = 5 SE +/- 0.057, N = 15 SE +/- 0.054, N = 15 SE +/- 0.012, N = 6 SE +/- 0.022, N = 7 SE +/- 0.063, N = 15 SE +/- 0.084, N = 15 8.541 8.572 4.723 4.491 6.662 5.804 3.780 3.361 1. (CC) gcc options: -lm -lpthread -O3 Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 8.49 / Avg: 8.54 / Max: 8.58 Min: 8.52 / Avg: 8.57 / Max: 8.64 Min: 4.37 / Avg: 4.72 / Max: 5.05 Min: 4.25 / Avg: 4.49 / Max: 4.93 Min: 6.62 / Avg: 6.66 / Max: 6.7 Min: 5.73 / Avg: 5.8 / Max: 5.9 Min: 3.49 / Avg: 3.78 / Max: 4.17 Min: 3.06 / Avg: 3.36 / Max: 4.06 1. (CC) gcc options: -lm -lpthread -O3
POV-Ray This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 SE +/- 0.037, N = 5 SE +/- 0.025, N = 5 SE +/- 0.035, N = 6 SE +/- 0.024, N = 6 SE +/- 0.034, N = 5 SE +/- 0.032, N = 5 SE +/- 0.041, N = 15 SE +/- 0.042, N = 15 7.445 7.145 5.851 5.733 7.178 6.578 6.035 5.673 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -R/usr/lib -lSDL -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 7.37 / Avg: 7.45 / Max: 7.56 Min: 7.05 / Avg: 7.15 / Max: 7.2 Min: 5.73 / Avg: 5.85 / Max: 5.99 Min: 5.63 / Avg: 5.73 / Max: 5.8 Min: 7.06 / Avg: 7.18 / Max: 7.24 Min: 6.47 / Avg: 6.58 / Max: 6.66 Min: 5.82 / Avg: 6.03 / Max: 6.39 Min: 5.41 / Avg: 5.67 / Max: 5.96 1. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -R/usr/lib -lSDL -lXpm -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
Embree Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs and supporting instruction sets such as SSE, AVX, AVX2, and AVX-512. Embree also supports making use of the Intel SPMD Program Compiler (ISPC). Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50 100 150 200 250 SE +/- 0.36, N = 6 SE +/- 0.12, N = 6 SE +/- 0.23, N = 8 SE +/- 0.18, N = 8 SE +/- 0.26, N = 7 SE +/- 0.14, N = 7 SE +/- 0.20, N = 8 SE +/- 0.29, N = 9 98.44 100.16 160.24 163.83 120.76 137.93 193.25 210.04 MIN: 94.48 / MAX: 100.85 MIN: 96.79 / MAX: 102 MIN: 154.06 / MAX: 165.16 MIN: 158.53 / MAX: 168.64 MIN: 116.51 / MAX: 124.3 MIN: 132.56 / MAX: 143.06 MIN: 188 / MAX: 207.13 MIN: 203.63 / MAX: 218.06 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Asian Dragon EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 Min: 97.33 / Avg: 98.44 / Max: 99.29 Min: 99.74 / Avg: 100.16 / Max: 100.41 Min: 159.24 / Avg: 160.24 / Max: 161.18 Min: 163.16 / Avg: 163.83 / Max: 164.46 Min: 119.42 / Avg: 120.76 / Max: 121.32 Min: 137.33 / Avg: 137.93 / Max: 138.47 Min: 192.39 / Avg: 193.25 / Max: 193.81 Min: 208.64 / Avg: 210.04 / Max: 211.08
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Crown EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.09, N = 6 SE +/- 0.09, N = 6 SE +/- 0.36, N = 7 SE +/- 0.31, N = 7 SE +/- 0.14, N = 6 SE +/- 0.08, N = 7 SE +/- 0.38, N = 8 SE +/- 0.43, N = 8 85.56 89.31 141.64 147.85 102.97 115.54 175.90 202.58 MIN: 83.63 / MAX: 88.64 MIN: 86.89 / MAX: 93.53 MIN: 122.02 / MAX: 159.72 MIN: 125.69 / MAX: 164.8 MIN: 97.6 / MAX: 109.75 MIN: 109.28 / MAX: 125.95 MIN: 132.24 / MAX: 198.5 MIN: 154.56 / MAX: 222.86 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer - Model: Crown EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 Min: 85.38 / Avg: 85.56 / Max: 85.97 Min: 89.01 / Avg: 89.31 / Max: 89.54 Min: 140.42 / Avg: 141.64 / Max: 142.98 Min: 146.81 / Avg: 147.85 / Max: 149.29 Min: 102.5 / Avg: 102.97 / Max: 103.46 Min: 115.27 / Avg: 115.54 / Max: 115.89 Min: 174.41 / Avg: 175.9 / Max: 177.95 Min: 200.62 / Avg: 202.58 / Max: 204.17
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50 100 150 200 250 SE +/- 0.15, N = 7 SE +/- 0.21, N = 7 SE +/- 0.39, N = 8 SE +/- 0.35, N = 8 SE +/- 0.07, N = 7 SE +/- 0.21, N = 8 SE +/- 0.22, N = 9 SE +/- 0.42, N = 9 111.70 116.10 176.35 178.97 136.11 154.64 213.25 232.40 MIN: 108.1 / MAX: 114.03 MIN: 111.56 / MAX: 118.75 MIN: 171.08 / MAX: 182.71 MIN: 174.04 / MAX: 186.85 MIN: 133.47 / MAX: 140.82 MIN: 149.61 / MAX: 162.58 MIN: 207.6 / MAX: 227.17 MIN: 224.07 / MAX: 241.37 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Asian Dragon EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 Min: 110.95 / Avg: 111.7 / Max: 112.24 Min: 114.88 / Avg: 116.1 / Max: 116.46 Min: 174.53 / Avg: 176.35 / Max: 177.6 Min: 177.68 / Avg: 178.97 / Max: 180.51 Min: 135.92 / Avg: 136.11 / Max: 136.41 Min: 153.64 / Avg: 154.64 / Max: 155.19 Min: 212.11 / Avg: 213.25 / Max: 214.12 Min: 229.25 / Avg: 232.4 / Max: 233.43
Result
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Crown EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50 100 150 200 250 SE +/- 0.09, N = 6 SE +/- 0.23, N = 6 SE +/- 0.35, N = 8 SE +/- 0.24, N = 8 SE +/- 0.16, N = 7 SE +/- 0.12, N = 7 SE +/- 0.38, N = 8 SE +/- 0.25, N = 8 91.72 96.79 149.02 157.17 109.92 123.62 182.48 212.17 MIN: 89.33 / MAX: 96.66 MIN: 93.34 / MAX: 102.27 MIN: 128.73 / MAX: 170.39 MIN: 131.7 / MAX: 178.28 MIN: 103.8 / MAX: 118.85 MIN: 116.19 / MAX: 135.91 MIN: 126.18 / MAX: 211.95 MIN: 156.18 / MAX: 236.74 Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Embree 3.13 Binary: Pathtracer ISPC - Model: Crown EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 Min: 91.43 / Avg: 91.72 / Max: 92.03 Min: 95.81 / Avg: 96.79 / Max: 97.26 Min: 147.94 / Avg: 149.02 / Max: 150.82 Min: 155.84 / Avg: 157.17 / Max: 157.95 Min: 109.54 / Avg: 109.92 / Max: 110.54 Min: 122.98 / Avg: 123.62 / Max: 123.97 Min: 181.2 / Avg: 182.48 / Max: 184.79 Min: 211.47 / Avg: 212.17 / Max: 213.44
Intel Open Image Denoise Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.8798 1.7596 2.6394 3.5192 4.399 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.02, N = 5 SE +/- 0.02, N = 5 2.67 2.73 3.84 3.91 3.04 3.33 3.57 3.70 Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.hdr_alb_nrm.3840x2160 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 2.67 / Avg: 2.67 / Max: 2.67 Min: 2.72 / Avg: 2.73 / Max: 2.73 Min: 3.8 / Avg: 3.84 / Max: 3.87 Min: 3.89 / Avg: 3.91 / Max: 3.93 Min: 3.04 / Avg: 3.04 / Max: 3.04 Min: 3.32 / Avg: 3.33 / Max: 3.34 Min: 3.52 / Avg: 3.57 / Max: 3.61 Min: 3.65 / Avg: 3.7 / Max: 3.74
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.ldr_alb_nrm.3840x2160 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.8753 1.7506 2.6259 3.5012 4.3765 SE +/- 0.00, N = 4 SE +/- 0.00, N = 4 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.00, N = 5 SE +/- 0.00, N = 5 SE +/- 0.02, N = 5 SE +/- 0.01, N = 5 2.67 2.73 3.82 3.89 3.05 3.33 3.57 3.69 Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RT.ldr_alb_nrm.3840x2160 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 2.67 / Avg: 2.67 / Max: 2.67 Min: 2.73 / Avg: 2.73 / Max: 2.73 Min: 3.78 / Avg: 3.82 / Max: 3.85 Min: 3.87 / Avg: 3.89 / Max: 3.9 Min: 3.04 / Avg: 3.05 / Max: 3.05 Min: 3.33 / Avg: 3.33 / Max: 3.34 Min: 3.53 / Avg: 3.57 / Max: 3.62 Min: 3.67 / Avg: 3.69 / Max: 3.73
Result
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.414 0.828 1.242 1.656 2.07 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 1.26 1.29 1.81 1.84 1.44 1.58 1.69 1.77 Result Confidence
OpenBenchmarking.org Images / Sec, More Is Better Intel Open Image Denoise 1.4.0 Run: RTLightmap.hdr.4096x4096 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 1.26 / Avg: 1.26 / Max: 1.26 Min: 1.29 / Avg: 1.29 / Max: 1.29 Min: 1.79 / Avg: 1.81 / Max: 1.81 Min: 1.84 / Avg: 1.84 / Max: 1.85 Min: 1.44 / Avg: 1.44 / Max: 1.44 Min: 1.58 / Avg: 1.58 / Max: 1.58 Min: 1.68 / Avg: 1.69 / Max: 1.7 Min: 1.75 / Avg: 1.77 / Max: 1.77
OSPRay Intel OSPRay is a portable ray-tracing engine for high-performance, high-fidelity scientific visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/ao/real_time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 12 24 36 48 60 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.18, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 21.52 22.02 37.05 37.95 26.53 30.56 43.96 51.71 Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/ao/real_time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 10 20 30 40 50 Min: 21.41 / Avg: 21.52 / Max: 21.58 Min: 22.01 / Avg: 22.02 / Max: 22.03 Min: 37.02 / Avg: 37.05 / Max: 37.1 Min: 37.85 / Avg: 37.95 / Max: 38 Min: 26.52 / Avg: 26.53 / Max: 26.55 Min: 30.19 / Avg: 30.56 / Max: 30.76 Min: 43.91 / Avg: 43.96 / Max: 44.04 Min: 51.61 / Avg: 51.71 / Max: 51.87
Result
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 14 28 42 56 70 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 25.70 26.45 46.03 48.06 31.17 36.32 54.21 64.14 Result Confidence
OpenBenchmarking.org Items Per Second, More Is Better OSPRay 2.10 Benchmark: gravity_spheres_volume/dim_512/pathtracer/real_time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 13 26 39 52 65 Min: 25.68 / Avg: 25.7 / Max: 25.72 Min: 26.42 / Avg: 26.45 / Max: 26.47 Min: 46 / Avg: 46.03 / Max: 46.06 Min: 48.01 / Avg: 48.06 / Max: 48.1 Min: 31.15 / Avg: 31.17 / Max: 31.21 Min: 36.28 / Avg: 36.32 / Max: 36.38 Min: 54.07 / Avg: 54.21 / Max: 54.4 Min: 63.97 / Avg: 64.14 / Max: 64.32
OSPRay Studio Intel OSPRay Studio is an open-source, interactive visualization and ray-tracing software package. OSPRay Studio makes use of Intel OSPRay, a portable ray-tracing engine for high-performance, high-fidelity visualizations. OSPRay builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 SE +/- 0.67, N = 3 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 1289 1240 717 667 1060 916 579 503 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 Min: 1288 / Avg: 1289.33 / Max: 1290 Min: 1239 / Avg: 1240.33 / Max: 1242 Min: 716 / Avg: 717.33 / Max: 718 Min: 666 / Avg: 666.67 / Max: 667 Min: 1059 / Avg: 1059.67 / Max: 1060 Min: 915 / Avg: 916.33 / Max: 918 Min: 578 / Avg: 579.33 / Max: 580 Min: 502 / Avg: 502.67 / Max: 503 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K SE +/- 7.54, N = 3 SE +/- 7.06, N = 3 SE +/- 13.62, N = 3 SE +/- 8.50, N = 3 SE +/- 11.92, N = 3 SE +/- 20.21, N = 3 SE +/- 6.17, N = 3 SE +/- 13.35, N = 3 20623 19898 11454 10640 16971 14722 9256 8043 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K Min: 20613 / Avg: 20623.33 / Max: 20638 Min: 19885 / Avg: 19898.33 / Max: 19909 Min: 11432 / Avg: 11454.33 / Max: 11479 Min: 10623 / Avg: 10640 / Max: 10649 Min: 16957 / Avg: 16971.33 / Max: 16995 Min: 14687 / Avg: 14722 / Max: 14757 Min: 9247 / Avg: 9256.33 / Max: 9268 Min: 8016 / Avg: 8042.67 / Max: 8057 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9K 18K 27K 36K 45K SE +/- 31.01, N = 3 SE +/- 55.41, N = 3 SE +/- 43.24, N = 3 SE +/- 33.62, N = 3 SE +/- 53.67, N = 3 SE +/- 57.52, N = 3 SE +/- 50.75, N = 3 SE +/- 32.54, N = 3 41316 39777 22971 21323 33978 29445 18567 16134 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 1 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 7K 14K 21K 28K 35K Min: 41277 / Avg: 41315.67 / Max: 41377 Min: 39671 / Avg: 39777 / Max: 39858 Min: 22909 / Avg: 22970.67 / Max: 23054 Min: 21280 / Avg: 21322.67 / Max: 21389 Min: 33903 / Avg: 33978 / Max: 34082 Min: 29357 / Avg: 29444.67 / Max: 29553 Min: 18484 / Avg: 18566.67 / Max: 18659 Min: 16089 / Avg: 16133.67 / Max: 16197 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 SE +/- 1.86, N = 3 SE +/- 1.53, N = 3 SE +/- 0.67, N = 3 SE +/- 1.15, N = 3 SE +/- 0.33, N = 3 SE +/- 0.67, N = 3 SE +/- 1.53, N = 3 SE +/- 1.53, N = 3 1310 1262 731 683 1071 926 600 510 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 2 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 Min: 1308 / Avg: 1310.33 / Max: 1314 Min: 1260 / Avg: 1262 / Max: 1265 Min: 730 / Avg: 731.33 / Max: 732 Min: 681 / Avg: 683 / Max: 685 Min: 1070 / Avg: 1070.67 / Max: 1071 Min: 925 / Avg: 925.67 / Max: 927 Min: 597 / Avg: 600 / Max: 602 Min: 508 / Avg: 510 / Max: 513 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K SE +/- 16.33, N = 3 SE +/- 14.57, N = 3 SE +/- 5.69, N = 3 SE +/- 12.91, N = 3 SE +/- 9.61, N = 3 SE +/- 40.04, N = 3 SE +/- 14.38, N = 3 SE +/- 11.98, N = 3 20946 20230 11646 10889 17162 14926 9375 8172 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 2 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K Min: 20915 / Avg: 20946.33 / Max: 20970 Min: 20201 / Avg: 20230 / Max: 20247 Min: 11635 / Avg: 11646 / Max: 11654 Min: 10874 / Avg: 10889.33 / Max: 10915 Min: 17149 / Avg: 17162.33 / Max: 17181 Min: 14869 / Avg: 14925.67 / Max: 15003 Min: 9346 / Avg: 9374.67 / Max: 9391 Min: 8150 / Avg: 8172.33 / Max: 8191 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9K 18K 27K 36K 45K SE +/- 34.87, N = 3 SE +/- 60.79, N = 3 SE +/- 24.88, N = 3 SE +/- 12.86, N = 3 SE +/- 75.68, N = 3 SE +/- 52.88, N = 3 SE +/- 47.26, N = 3 SE +/- 9.49, N = 3 41738 40406 23309 21776 34323 29715 18823 16356 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 2 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 7K 14K 21K 28K 35K Min: 41674 / Avg: 41738 / Max: 41794 Min: 40314 / Avg: 40406.33 / Max: 40521 Min: 23276 / Avg: 23309.33 / Max: 23358 Min: 21752 / Avg: 21776 / Max: 21796 Min: 34219 / Avg: 34322.67 / Max: 34470 Min: 29618 / Avg: 29715 / Max: 29800 Min: 18731 / Avg: 18823.33 / Max: 18887 Min: 16344 / Avg: 16356.33 / Max: 16375 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 SE +/- 1.15, N = 3 SE +/- 1.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 1.20, N = 3 SE +/- 1.45, N = 3 SE +/- 1.76, N = 3 1537 1483 859 799 1263 1095 693 603 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 1 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 Min: 1535 / Avg: 1537 / Max: 1539 Min: 1480 / Avg: 1482.67 / Max: 1484 Min: 857 / Avg: 858.67 / Max: 860 Min: 798 / Avg: 799 / Max: 800 Min: 1262 / Avg: 1263 / Max: 1264 Min: 1093 / Avg: 1094.67 / Max: 1097 Min: 690 / Avg: 692.67 / Max: 695 Min: 600 / Avg: 602.67 / Max: 606 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5K 10K 15K 20K 25K SE +/- 19.97, N = 3 SE +/- 8.19, N = 3 SE +/- 14.85, N = 3 SE +/- 3.06, N = 3 SE +/- 5.61, N = 3 SE +/- 30.02, N = 3 SE +/- 16.26, N = 3 SE +/- 10.97, N = 3 24607 23782 13690 12789 20228 17569 11055 9597 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 16 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4K 8K 12K 16K 20K Min: 24577 / Avg: 24607.33 / Max: 24645 Min: 23767 / Avg: 23782.33 / Max: 23795 Min: 13661 / Avg: 13690.33 / Max: 13709 Min: 12785 / Avg: 12789 / Max: 12795 Min: 20217 / Avg: 20227.67 / Max: 20236 Min: 17531 / Avg: 17568.67 / Max: 17628 Min: 11034 / Avg: 11055 / Max: 11087 Min: 9578 / Avg: 9596.67 / Max: 9616 1. (CXX) g++ options: -O3 -ldl
Result
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 11K 22K 33K 44K 55K SE +/- 30.66, N = 3 SE +/- 21.43, N = 3 SE +/- 32.90, N = 3 SE +/- 30.95, N = 3 SE +/- 80.97, N = 3 SE +/- 70.57, N = 3 SE +/- 20.98, N = 3 SE +/- 26.91, N = 3 49152 47638 27438 25561 40492 35059 22083 19215 1. (CXX) g++ options: -O3 -ldl Result Confidence
OpenBenchmarking.org ms, Fewer Is Better OSPRay Studio 0.11 Camera: 3 - Resolution: 4K - Samples Per Pixel: 32 - Renderer: Path Tracer EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9K 18K 27K 36K 45K Min: 49112 / Avg: 49151.67 / Max: 49212 Min: 47603 / Avg: 47638.33 / Max: 47677 Min: 27372 / Avg: 27437.67 / Max: 27474 Min: 25503 / Avg: 25560.67 / Max: 25609 Min: 40338 / Avg: 40492.33 / Max: 40612 Min: 34954 / Avg: 35058.67 / Max: 35193 Min: 22043 / Avg: 22083 / Max: 22114 Min: 19163 / Avg: 19215 / Max: 19253 1. (CXX) g++ options: -O3 -ldl
LuxCoreRender LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 6 12 18 24 30 SE +/- 0.05, N = 3 SE +/- 0.11, N = 10 SE +/- 0.47, N = 15 SE +/- 0.15, N = 3 SE +/- 0.02, N = 3 SE +/- 0.21, N = 15 SE +/- 0.07, N = 3 SE +/- 1.00, N = 15 12.86 13.60 17.68 21.53 14.31 17.07 17.66 23.01 MIN: 12.51 / MAX: 14.41 MIN: 13.01 / MAX: 15.51 MIN: 15.4 / MAX: 25.14 MIN: 20.88 / MAX: 26.15 MIN: 13.83 / MAX: 16.93 MIN: 15.77 / MAX: 20.52 MIN: 16.23 / MAX: 28.26 MIN: 17.54 / MAX: 34.17 Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: DLSC - Acceleration: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 Min: 12.77 / Avg: 12.86 / Max: 12.94 Min: 13.36 / Avg: 13.6 / Max: 14.27 Min: 16.31 / Avg: 17.68 / Max: 21.26 Min: 21.36 / Avg: 21.53 / Max: 21.83 Min: 14.28 / Avg: 14.31 / Max: 14.33 Min: 16.41 / Avg: 17.07 / Max: 18.44 Min: 17.57 / Avg: 17.66 / Max: 17.8 Min: 19.1 / Avg: 23.01 / Max: 27.17
Result
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 7 14 21 28 35 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 SE +/- 0.09, N = 3 SE +/- 0.57, N = 15 SE +/- 0.15, N = 15 SE +/- 0.02, N = 3 SE +/- 0.73, N = 15 SE +/- 0.19, N = 3 18.50 18.63 25.82 26.85 21.49 23.62 29.13 30.79 MIN: 16.15 / MAX: 22.48 MIN: 16.54 / MAX: 23.87 MIN: 21.62 / MAX: 38.24 MIN: 21.39 / MAX: 39.12 MIN: 18.29 / MAX: 27.17 MIN: 21.1 / MAX: 31.71 MIN: 22.84 / MAX: 45.75 MIN: 24.95 / MAX: 52.26 Result Confidence
OpenBenchmarking.org M samples/sec, More Is Better LuxCoreRender 2.6 Scene: Orange Juice - Acceleration: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 7 14 21 28 35 Min: 18.42 / Avg: 18.5 / Max: 18.62 Min: 18.54 / Avg: 18.63 / Max: 18.69 Min: 25.63 / Avg: 25.82 / Max: 25.94 Min: 25.17 / Avg: 26.85 / Max: 30.67 Min: 21.05 / Avg: 21.49 / Max: 22.74 Min: 23.6 / Avg: 23.62 / Max: 23.66 Min: 26.89 / Avg: 29.13 / Max: 34.92 Min: 30.54 / Avg: 30.79 / Max: 31.15
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Very Fast EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.20, N = 5 SE +/- 0.12, N = 5 SE +/- 0.40, N = 6 SE +/- 0.31, N = 6 SE +/- 0.51, N = 6 SE +/- 0.53, N = 15 SE +/- 0.55, N = 5 SE +/- 0.30, N = 5 62.33 62.13 73.37 74.70 76.67 77.05 71.57 71.57 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt Result Confidence
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Very Fast EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 15 30 45 60 75 Min: 61.55 / Avg: 62.33 / Max: 62.63 Min: 61.83 / Avg: 62.13 / Max: 62.46 Min: 72.01 / Avg: 73.37 / Max: 74.88 Min: 73.62 / Avg: 74.7 / Max: 75.8 Min: 74.23 / Avg: 76.67 / Max: 77.58 Min: 69.86 / Avg: 77.05 / Max: 78.24 Min: 70.29 / Avg: 71.57 / Max: 72.9 Min: 70.62 / Avg: 71.57 / Max: 72.26 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests on a sample 6000x4000 pixel JPEG image. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Enhanced EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 400 800 1200 1600 2000 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 SE +/- 1.45, N = 3 SE +/- 4.51, N = 3 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 8.96, N = 3 SE +/- 14.97, N = 3 1216 1239 1756 1737 1429 1665 1923 2080 1. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread Result Confidence
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Enhanced EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 400 800 1200 1600 2000 Min: 1215 / Avg: 1216 / Max: 1217 Min: 1238 / Avg: 1239.33 / Max: 1240 Min: 1754 / Avg: 1756.33 / Max: 1759 Min: 1732 / Avg: 1737 / Max: 1746 Min: 1428 / Avg: 1429 / Max: 1430 Min: 1665 / Avg: 1665.33 / Max: 1666 Min: 1907 / Avg: 1923 / Max: 1938 Min: 2050 / Avg: 2079.67 / Max: 2098 1. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
Result
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Sharpen EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 300 600 900 1200 1500 SE +/- 0.33, N = 3 SE +/- 0.33, N = 3 SE +/- 3.18, N = 3 SE +/- 1.67, N = 3 SE +/- 0.58, N = 3 SE +/- 0.67, N = 3 SE +/- 3.46, N = 3 SE +/- 2.91, N = 3 697 704 1102 1114 843 999 1183 1393 1. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread Result Confidence
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.38 Operation: Sharpen EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 Min: 697 / Avg: 697.33 / Max: 698 Min: 703 / Avg: 703.67 / Max: 704 Min: 1096 / Avg: 1101.67 / Max: 1107 Min: 1112 / Avg: 1113.67 / Max: 1117 Min: 842 / Avg: 843 / Max: 844 Min: 998 / Avg: 998.67 / Max: 1000 Min: 1177 / Avg: 1183 / Max: 1189 Min: 1388 / Avg: 1393.33 / Max: 1398 1. (CC) gcc options: -fopenmp -O2 -ljbig -lwebp -lwebpmux -ltiff -lfreetype -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lxml2 -lz -lm -lpthread
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Medium EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 130 260 390 520 650 SE +/- 0.06, N = 8 SE +/- 0.07, N = 8 SE +/- 0.91, N = 8 SE +/- 0.91, N = 8 SE +/- 0.19, N = 8 SE +/- 0.34, N = 8 SE +/- 1.52, N = 8 SE +/- 1.79, N = 8 360.55 388.57 548.85 613.64 422.02 480.19 571.61 601.36 1. (CXX) g++ options: -O3 -flto -pthread Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Medium EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 110 220 330 440 550 Min: 360.25 / Avg: 360.55 / Max: 360.77 Min: 388.26 / Avg: 388.57 / Max: 388.89 Min: 543.61 / Avg: 548.85 / Max: 552.57 Min: 610.21 / Avg: 613.64 / Max: 617.71 Min: 421.42 / Avg: 422.02 / Max: 422.75 Min: 479.05 / Avg: 480.19 / Max: 481.42 Min: 568.31 / Avg: 571.61 / Max: 581.27 Min: 590.14 / Avg: 601.36 / Max: 606.6 1. (CXX) g++ options: -O3 -flto -pthread
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Thorough EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 SE +/- 0.01, N = 5 SE +/- 0.01, N = 5 SE +/- 0.01, N = 6 SE +/- 0.02, N = 7 SE +/- 0.01, N = 6 SE +/- 0.04, N = 6 SE +/- 0.05, N = 7 SE +/- 0.08, N = 7 47.54 50.94 90.55 98.40 56.20 65.40 106.62 122.12 1. (CXX) g++ options: -O3 -flto -pthread Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Thorough EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 47.49 / Avg: 47.54 / Max: 47.56 Min: 50.91 / Avg: 50.94 / Max: 50.97 Min: 90.52 / Avg: 90.55 / Max: 90.59 Min: 98.34 / Avg: 98.4 / Max: 98.47 Min: 56.15 / Avg: 56.2 / Max: 56.24 Min: 65.27 / Avg: 65.4 / Max: 65.53 Min: 106.39 / Avg: 106.62 / Max: 106.75 Min: 121.78 / Avg: 122.12 / Max: 122.39 1. (CXX) g++ options: -O3 -flto -pthread
Result
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Exhaustive EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 SE +/- 0.0005, N = 4 SE +/- 0.0021, N = 4 SE +/- 0.0052, N = 5 SE +/- 0.0010, N = 6 SE +/- 0.0016, N = 4 SE +/- 0.0031, N = 5 SE +/- 0.0146, N = 6 SE +/- 0.0373, N = 6 5.0295 5.3824 9.6418 10.7069 5.9568 7.0368 11.7394 13.6998 1. (CXX) g++ options: -O3 -flto -pthread Result Confidence
OpenBenchmarking.org MT/s, More Is Better ASTC Encoder 4.0 Preset: Exhaustive EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 Min: 5.03 / Avg: 5.03 / Max: 5.03 Min: 5.38 / Avg: 5.38 / Max: 5.39 Min: 9.63 / Avg: 9.64 / Max: 9.65 Min: 10.7 / Avg: 10.71 / Max: 10.71 Min: 5.95 / Avg: 5.96 / Max: 5.96 Min: 7.03 / Avg: 7.04 / Max: 7.04 Min: 11.72 / Avg: 11.74 / Max: 11.81 Min: 13.61 / Avg: 13.7 / Max: 13.84 1. (CXX) g++ options: -O3 -flto -pthread
Etcpak Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. The test profile uses a 8K x 8K game texture as a sample input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 1.0 Benchmark: Multi-Threaded - Configuration: ETC2 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 16.14, N = 9 SE +/- 18.91, N = 9 SE +/- 26.45, N = 9 SE +/- 28.22, N = 9 SE +/- 97.61, N = 15 SE +/- 79.32, N = 15 SE +/- 70.12, N = 9 SE +/- 47.66, N = 9 8966.41 8976.89 8902.85 8825.79 9257.88 9335.76 9305.67 9232.75 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread Result Confidence
OpenBenchmarking.org Mpx/s, More Is Better Etcpak 1.0 Benchmark: Multi-Threaded - Configuration: ETC2 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1600 3200 4800 6400 8000 Min: 8918.12 / Avg: 8966.41 / Max: 9068.77 Min: 8927.61 / Avg: 8976.89 / Max: 9108.15 Min: 8785.03 / Avg: 8902.85 / Max: 8995.83 Min: 8738.13 / Avg: 8825.79 / Max: 9026.07 Min: 8902.74 / Avg: 9257.88 / Max: 10138.82 Min: 8916.94 / Avg: 9335.76 / Max: 9914.15 Min: 8935.93 / Avg: 9305.67 / Max: 9649.01 Min: 9050.42 / Avg: 9232.75 / Max: 9489.38 1. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
ebizzy This is a test of ebizzy, a program to generate workloads resembling web server workloads. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1.2M 2.4M 3.6M 4.8M 6M SE +/- 41834.71, N = 15 SE +/- 38422.79, N = 13 SE +/- 108149.58, N = 15 SE +/- 105603.89, N = 15 SE +/- 56812.23, N = 15 SE +/- 53379.81, N = 15 SE +/- 133651.04, N = 12 SE +/- 121230.32, N = 15 4013193 4769106 5393882 5820948 4287200 4670562 5250316 5731049 1. (CC) gcc options: -pthread -lpthread -O3 -march=native Result Confidence
OpenBenchmarking.org Records/s, More Is Better ebizzy 0.3 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1000K 2000K 3000K 4000K 5000K Min: 3753846 / Avg: 4013193.07 / Max: 4313815 Min: 4410437 / Avg: 4769105.85 / Max: 4915642 Min: 4781811 / Avg: 5393881.53 / Max: 6070632 Min: 5077027 / Avg: 5820948.07 / Max: 6346253 Min: 3989540 / Avg: 4287200 / Max: 4694682 Min: 4375248 / Avg: 4670561.6 / Max: 5026047 Min: 4524510 / Avg: 5250316.25 / Max: 6091132 Min: 4876308 / Avg: 5731048.73 / Max: 6551636 1. (CC) gcc options: -pthread -lpthread -O3 -march=native
Node.js Express HTTP Load Test A Node.js Express server with a Node-based loadtest client for facilitating HTTP benchmarking. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2K 4K 6K 8K 10K SE +/- 39.98, N = 4 SE +/- 27.20, N = 4 SE +/- 36.28, N = 4 SE +/- 74.87, N = 5 SE +/- 11.05, N = 4 SE +/- 17.64, N = 4 SE +/- 62.92, N = 8 SE +/- 73.54, N = 6 7846 7884 7453 7500 7829 7780 7508 7474 Result Confidence
OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1400 2800 4200 5600 7000 Min: 7773 / Avg: 7846 / Max: 7937 Min: 7831 / Avg: 7884 / Max: 7959 Min: 7397 / Avg: 7452.5 / Max: 7559 Min: 7252 / Avg: 7500.2 / Max: 7685 Min: 7806 / Avg: 7829.25 / Max: 7851 Min: 7747 / Avg: 7780 / Max: 7811 Min: 7216 / Avg: 7508 / Max: 7731 Min: 7132 / Avg: 7473.5 / Max: 7666
simdjson This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: PartialTweets EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1.287 2.574 3.861 5.148 6.435 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 5.70 5.72 5.70 5.72 5.67 5.66 5.67 5.67 1. (CXX) g++ options: -O3 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: PartialTweets EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 5.68 / Avg: 5.7 / Max: 5.73 Min: 5.69 / Avg: 5.72 / Max: 5.74 Min: 5.68 / Avg: 5.7 / Max: 5.72 Min: 5.68 / Avg: 5.72 / Max: 5.76 Min: 5.64 / Avg: 5.67 / Max: 5.69 Min: 5.65 / Avg: 5.66 / Max: 5.68 Min: 5.65 / Avg: 5.67 / Max: 5.7 Min: 5.63 / Avg: 5.67 / Max: 5.69 1. (CXX) g++ options: -O3
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: LargeRandom EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.2858 0.5716 0.8574 1.1432 1.429 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 1.27 1.26 1.27 1.27 1.25 1.25 1.25 1.25 1. (CXX) g++ options: -O3 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: LargeRandom EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 1.27 / Avg: 1.27 / Max: 1.27 Min: 1.26 / Avg: 1.26 / Max: 1.26 Min: 1.27 / Avg: 1.27 / Max: 1.27 Min: 1.27 / Avg: 1.27 / Max: 1.27 Min: 1.25 / Avg: 1.25 / Max: 1.25 Min: 1.25 / Avg: 1.25 / Max: 1.25 Min: 1.25 / Avg: 1.25 / Max: 1.25 Min: 1.25 / Avg: 1.25 / Max: 1.25 1. (CXX) g++ options: -O3
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: Kostya EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.9383 1.8766 2.8149 3.7532 4.6915 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 4.17 4.17 4.15 4.16 4.10 4.11 4.10 4.11 1. (CXX) g++ options: -O3 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: Kostya EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 4.17 / Avg: 4.17 / Max: 4.17 Min: 4.16 / Avg: 4.17 / Max: 4.18 Min: 4.13 / Avg: 4.15 / Max: 4.16 Min: 4.16 / Avg: 4.16 / Max: 4.16 Min: 4.09 / Avg: 4.1 / Max: 4.11 Min: 4.11 / Avg: 4.11 / Max: 4.12 Min: 4.08 / Avg: 4.1 / Max: 4.11 Min: 4.1 / Avg: 4.11 / Max: 4.11 1. (CXX) g++ options: -O3
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: DistinctUserID EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 6.95 6.91 6.92 6.93 6.82 6.83 6.86 6.87 1. (CXX) g++ options: -O3 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: DistinctUserID EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 6.92 / Avg: 6.95 / Max: 6.97 Min: 6.89 / Avg: 6.91 / Max: 6.94 Min: 6.85 / Avg: 6.92 / Max: 6.96 Min: 6.88 / Avg: 6.93 / Max: 6.98 Min: 6.8 / Avg: 6.82 / Max: 6.83 Min: 6.78 / Avg: 6.83 / Max: 6.87 Min: 6.84 / Avg: 6.86 / Max: 6.88 Min: 6.84 / Avg: 6.87 / Max: 6.9 1. (CXX) g++ options: -O3
Result
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: TopTweet EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 6.65 6.60 6.66 6.65 6.57 6.55 6.49 6.57 1. (CXX) g++ options: -O3 Result Confidence
OpenBenchmarking.org GB/s, More Is Better simdjson 2.0 Throughput Test: TopTweet EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 6.61 / Avg: 6.65 / Max: 6.68 Min: 6.51 / Avg: 6.6 / Max: 6.66 Min: 6.65 / Avg: 6.66 / Max: 6.67 Min: 6.6 / Avg: 6.65 / Max: 6.69 Min: 6.54 / Avg: 6.57 / Max: 6.59 Min: 6.49 / Avg: 6.55 / Max: 6.59 Min: 6.41 / Avg: 6.49 / Max: 6.55 Min: 6.55 / Avg: 6.57 / Max: 6.6 1. (CXX) g++ options: -O3
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 7.32.6 VGR Performance Metric EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1.2M 2.4M 3.6M 4.8M 6M 1196375 1247960 4006667 4136784 3123990 3179457 5793750 5820929 1. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -ldl -lm
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200 400 600 800 1000 SE +/- 3.84, N = 3 SE +/- 1.15, N = 3 SE +/- 2.33, N = 3 SE +/- 3.93, N = 3 SE +/- 6.01, N = 3 SE +/- 6.08, N = 3 SE +/- 5.61, N = 3 SE +/- 4.33, N = 3 849 849 844 846 855 864 864 862 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 150 300 450 600 750 Min: 843 / Avg: 848.67 / Max: 856 Min: 847 / Avg: 849 / Max: 851 Min: 839 / Avg: 843.67 / Max: 846 Min: 841 / Avg: 846.33 / Max: 854 Min: 843 / Avg: 854.67 / Max: 863 Min: 853 / Avg: 864 / Max: 874 Min: 853 / Avg: 863.67 / Max: 872 Min: 855 / Avg: 862.33 / Max: 870
PyPerformance PyPerformance is the reference Python performance benchmark suite. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: crypto_pyaes EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.17, N = 3 SE +/- 0.26, N = 3 SE +/- 0.20, N = 3 SE +/- 0.12, N = 3 SE +/- 0.21, N = 3 SE +/- 0.15, N = 3 SE +/- 0.12, N = 3 SE +/- 0.27, N = 3 87.0 86.8 86.6 87.3 88.2 87.4 90.1 89.5 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: crypto_pyaes EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 86.7 / Avg: 87.03 / Max: 87.2 Min: 86.3 / Avg: 86.8 / Max: 87.2 Min: 86.2 / Avg: 86.57 / Max: 86.9 Min: 87.1 / Avg: 87.3 / Max: 87.5 Min: 87.9 / Avg: 88.2 / Max: 88.6 Min: 87.2 / Avg: 87.43 / Max: 87.7 Min: 89.9 / Avg: 90.07 / Max: 90.3 Min: 89.1 / Avg: 89.47 / Max: 90
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: django_template EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 36.0 36.0 35.8 35.7 36.3 36.5 36.1 36.2 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: django_template EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 8 16 24 32 40 Min: 35.9 / Avg: 35.97 / Max: 36 Min: 35.9 / Avg: 35.97 / Max: 36 Min: 35.7 / Avg: 35.8 / Max: 35.9 Min: 35.6 / Avg: 35.73 / Max: 35.9 Min: 36.2 / Avg: 36.3 / Max: 36.4 Min: 36.4 / Avg: 36.53 / Max: 36.6 Min: 36.1 / Avg: 36.13 / Max: 36.2 Min: 36.2 / Avg: 36.23 / Max: 36.3
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: json_loads EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 16.3 16.3 16.5 16.4 16.6 16.7 16.5 16.6 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: json_loads EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 Min: 16.3 / Avg: 16.3 / Max: 16.3 Min: 16.3 / Avg: 16.33 / Max: 16.4 Min: 16.4 / Avg: 16.47 / Max: 16.5 Min: 16.4 / Avg: 16.4 / Max: 16.4 Min: 16.5 / Avg: 16.57 / Max: 16.6 Min: 16.6 / Avg: 16.67 / Max: 16.7 Min: 16.5 / Avg: 16.5 / Max: 16.5 Min: 16.6 / Avg: 16.6 / Max: 16.6
Result
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: regex_compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 115 114 114 115 116 116 116 117 Result Confidence
OpenBenchmarking.org Milliseconds, Fewer Is Better PyPerformance 1.0.0 Benchmark: regex_compile EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 115 / Avg: 115 / Max: 115 Min: 114 / Avg: 114 / Max: 114 Min: 114 / Avg: 114 / Max: 114 Min: 115 / Avg: 115 / Max: 115 Min: 116 / Avg: 116 / Max: 116 Min: 116 / Avg: 116 / Max: 116 Min: 116 / Avg: 116 / Max: 116 Min: 117 / Avg: 117 / Max: 117
Numpy Benchmark This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better Numpy Benchmark EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 130 260 390 520 650 SE +/- 0.88, N = 3 SE +/- 0.79, N = 3 SE +/- 1.62, N = 3 SE +/- 2.27, N = 3 SE +/- 1.13, N = 3 SE +/- 0.08, N = 3 SE +/- 1.32, N = 3 SE +/- 1.36, N = 3 592.89 593.13 592.13 591.08 585.06 583.50 586.86 584.26 Result Confidence
OpenBenchmarking.org Score, More Is Better Numpy Benchmark EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 100 200 300 400 500 Min: 591.17 / Avg: 592.89 / Max: 594.09 Min: 591.85 / Avg: 593.13 / Max: 594.58 Min: 588.98 / Avg: 592.13 / Max: 594.34 Min: 587.78 / Avg: 591.08 / Max: 595.44 Min: 582.96 / Avg: 585.06 / Max: 586.84 Min: 583.37 / Avg: 583.5 / Max: 583.65 Min: 584.24 / Avg: 586.86 / Max: 588.49 Min: 582.74 / Avg: 584.26 / Max: 586.97
PHPBench PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200K 400K 600K 800K 1000K SE +/- 4409.84, N = 3 SE +/- 9166.50, N = 3 SE +/- 9437.20, N = 4 SE +/- 1751.73, N = 3 SE +/- 5980.59, N = 3 SE +/- 1659.67, N = 3 SE +/- 7390.85, N = 3 SE +/- 3619.22, N = 3 855637 862102 876214 849730 843665 843411 847129 846790 Result Confidence
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 150K 300K 450K 600K 750K Min: 847369 / Avg: 855637 / Max: 862430 Min: 849425 / Avg: 862102.33 / Max: 879910 Min: 855313 / Avg: 876213.75 / Max: 900582 Min: 846404 / Avg: 849730.33 / Max: 852346 Min: 832206 / Avg: 843665.33 / Max: 852364 Min: 840328 / Avg: 843410.67 / Max: 846018 Min: 838035 / Avg: 847129 / Max: 861768 Min: 843102 / Avg: 846790 / Max: 854028
srsRAN srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN 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.
Result
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 90 180 270 360 450 SE +/- 0.42, N = 4 SE +/- 1.88, N = 4 SE +/- 1.84, N = 4 SE +/- 0.67, N = 4 SE +/- 0.30, N = 4 SE +/- 1.45, N = 4 SE +/- 0.49, N = 4 SE +/- 0.79, N = 4 421.2 420.0 419.5 421.6 416.8 413.3 416.7 414.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 70 140 210 280 350 Min: 420.4 / Avg: 421.15 / Max: 422.1 Min: 414.4 / Avg: 420.03 / Max: 422.1 Min: 414.2 / Avg: 419.45 / Max: 422.3 Min: 419.6 / Avg: 421.58 / Max: 422.6 Min: 416 / Avg: 416.83 / Max: 417.4 Min: 410.1 / Avg: 413.3 / Max: 416.9 Min: 415.3 / Avg: 416.7 / Max: 417.4 Min: 412.5 / Avg: 414.8 / Max: 416.1 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.68, N = 4 SE +/- 0.65, N = 4 SE +/- 0.81, N = 4 SE +/- 0.57, N = 4 SE +/- 0.15, N = 4 SE +/- 0.38, N = 4 SE +/- 0.84, N = 4 SE +/- 0.36, N = 4 168.1 169.0 168.0 168.7 166.6 166.6 165.2 166.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 166.1 / Avg: 168.08 / Max: 169.2 Min: 167.4 / Avg: 169 / Max: 170.5 Min: 166.2 / Avg: 168 / Max: 169.6 Min: 167.2 / Avg: 168.73 / Max: 169.9 Min: 166.3 / Avg: 166.63 / Max: 167 Min: 165.5 / Avg: 166.58 / Max: 167.2 Min: 163.2 / Avg: 165.23 / Max: 167.3 Min: 165.6 / Avg: 166.33 / Max: 167.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.81, N = 3 SE +/- 0.66, N = 3 SE +/- 0.54, N = 3 SE +/- 0.41, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 SE +/- 0.51, N = 3 SE +/- 0.23, N = 3 175.1 175.6 174.0 175.5 172.8 172.1 173.2 173.1 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB SISO 256-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 174.1 / Avg: 175.1 / Max: 176.7 Min: 174.4 / Avg: 175.57 / Max: 176.7 Min: 173.4 / Avg: 174.03 / Max: 175.1 Min: 174.8 / Avg: 175.47 / Max: 176.2 Min: 171.7 / Avg: 172.77 / Max: 173.7 Min: 170.3 / Avg: 172.07 / Max: 173 Min: 172.5 / Avg: 173.2 / Max: 174.2 Min: 172.6 / Avg: 173.07 / Max: 173.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 90 180 270 360 450 SE +/- 3.14, N = 3 SE +/- 0.13, N = 3 SE +/- 1.03, N = 3 SE +/- 0.90, N = 3 SE +/- 1.11, N = 3 SE +/- 2.63, N = 14 SE +/- 0.93, N = 3 SE +/- 0.25, N = 3 418.4 422.8 420.9 420.2 415.7 411.7 416.5 415.0 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 80 160 240 320 400 Min: 412.1 / Avg: 418.37 / Max: 421.9 Min: 422.7 / Avg: 422.83 / Max: 423.1 Min: 418.8 / Avg: 420.87 / Max: 421.9 Min: 418.5 / Avg: 420.23 / Max: 421.5 Min: 413.5 / Avg: 415.67 / Max: 417.2 Min: 379.3 / Avg: 411.73 / Max: 416.6 Min: 414.7 / Avg: 416.53 / Max: 417.7 Min: 414.7 / Avg: 415 / Max: 415.5 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.03, N = 3 SE +/- 0.43, N = 3 SE +/- 0.35, N = 3 SE +/- 0.27, N = 3 SE +/- 0.19, N = 3 SE +/- 0.20, N = 14 SE +/- 0.12, N = 3 SE +/- 0.15, N = 3 160.4 160.9 160.1 159.6 158.5 158.4 158.6 158.1 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 160.4 / Avg: 160.43 / Max: 160.5 Min: 160.1 / Avg: 160.93 / Max: 161.5 Min: 159.7 / Avg: 160.1 / Max: 160.8 Min: 159.2 / Avg: 159.57 / Max: 160.1 Min: 158.1 / Avg: 158.47 / Max: 158.7 Min: 157 / Avg: 158.36 / Max: 159.4 Min: 158.4 / Avg: 158.63 / Max: 158.8 Min: 157.8 / Avg: 158.1 / Max: 158.3 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 100 200 300 400 500 SE +/- 1.95, N = 3 SE +/- 2.79, N = 3 SE +/- 2.30, N = 3 SE +/- 0.78, N = 3 SE +/- 3.64, N = 3 SE +/- 6.35, N = 3 SE +/- 1.24, N = 3 SE +/- 2.00, N = 3 451.3 448.7 450.7 452.9 442.0 441.3 445.4 444.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 80 160 240 320 400 Min: 447.4 / Avg: 451.27 / Max: 453.6 Min: 445 / Avg: 448.73 / Max: 454.2 Min: 446.1 / Avg: 450.7 / Max: 453.1 Min: 451.9 / Avg: 452.87 / Max: 454.4 Min: 435.3 / Avg: 442.03 / Max: 447.8 Min: 428.6 / Avg: 441.3 / Max: 447.7 Min: 443.3 / Avg: 445.37 / Max: 447.6 Min: 440.8 / Avg: 444.8 / Max: 446.9 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40 80 120 160 200 SE +/- 0.13, N = 3 SE +/- 0.23, N = 3 SE +/- 0.26, N = 3 SE +/- 0.00, N = 3 SE +/- 0.54, N = 3 SE +/- 0.29, N = 3 SE +/- 0.19, N = 3 SE +/- 0.58, N = 3 169.3 169.1 168.5 168.9 166.2 166.4 166.1 166.7 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 169.2 / Avg: 169.33 / Max: 169.6 Min: 168.7 / Avg: 169.13 / Max: 169.5 Min: 168 / Avg: 168.47 / Max: 168.9 Min: 168.9 / Avg: 168.9 / Max: 168.9 Min: 165.6 / Avg: 166.23 / Max: 167.3 Min: 165.9 / Avg: 166.43 / Max: 166.9 Min: 165.7 / Avg: 166.07 / Max: 166.3 Min: 165.7 / Avg: 166.67 / Max: 167.7 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 SE +/- 0.06, N = 4 SE +/- 0.45, N = 4 SE +/- 0.23, N = 4 SE +/- 0.44, N = 4 SE +/- 0.34, N = 4 SE +/- 0.42, N = 4 SE +/- 0.44, N = 4 SE +/- 0.40, N = 4 141.9 141.6 141.7 141.0 139.4 139.8 139.3 138.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org eNb Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30 60 90 120 150 Min: 141.8 / Avg: 141.93 / Max: 142.1 Min: 140.3 / Avg: 141.63 / Max: 142.3 Min: 141 / Avg: 141.65 / Max: 142 Min: 140.2 / Avg: 141 / Max: 142.2 Min: 138.5 / Avg: 139.43 / Max: 140 Min: 138.5 / Avg: 139.75 / Max: 140.3 Min: 138.2 / Avg: 139.3 / Max: 140 Min: 138.1 / Avg: 138.8 / Max: 139.8 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 SE +/- 0.28, N = 4 SE +/- 0.29, N = 4 SE +/- 2.78, N = 4 SE +/- 0.36, N = 4 SE +/- 2.60, N = 4 SE +/- 0.09, N = 4 SE +/- 0.17, N = 4 SE +/- 0.31, N = 4 95.9 95.9 93.4 95.6 92.2 94.5 94.9 94.4 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org UE Mb/s, More Is Better srsRAN 22.04.1 Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAM EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20 40 60 80 100 Min: 95.2 / Avg: 95.9 / Max: 96.4 Min: 95.1 / Avg: 95.88 / Max: 96.4 Min: 85.1 / Avg: 93.43 / Max: 96.3 Min: 94.7 / Avg: 95.63 / Max: 96.2 Min: 84.4 / Avg: 92.2 / Max: 95.2 Min: 94.3 / Avg: 94.53 / Max: 94.7 Min: 94.4 / Avg: 94.9 / Max: 95.2 Min: 93.5 / Avg: 94.38 / Max: 94.9 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
Result
OpenBenchmarking.org Samples / Second, More Is Better srsRAN 22.04.1 Test: OFDM_Test EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 40M 80M 120M 160M 200M SE +/- 592546.29, N = 3 SE +/- 550757.05, N = 3 SE +/- 688799.28, N = 3 SE +/- 1211518.79, N = 3 SE +/- 2007485.99, N = 3 SE +/- 1212530.99, N = 11 SE +/- 1841496.98, N = 3 SE +/- 520683.31, N = 3 164566667 166800000 166666667 164166667 162000000 161236364 158633333 163766667 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm Result Confidence
OpenBenchmarking.org Samples / Second, More Is Better srsRAN 22.04.1 Test: OFDM_Test EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30M 60M 90M 120M 150M Min: 163700000 / Avg: 164566666.67 / Max: 165700000 Min: 165900000 / Avg: 166800000 / Max: 167800000 Min: 165300000 / Avg: 166666666.67 / Max: 167500000 Min: 161800000 / Avg: 164166666.67 / Max: 165800000 Min: 158300000 / Avg: 162000000 / Max: 165200000 Min: 152600000 / Avg: 161236363.64 / Max: 165400000 Min: 156500000 / Avg: 158633333.33 / Max: 162300000 Min: 162900000 / Avg: 163766666.67 / Max: 164700000 1. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -mavx512f -mavx512cd -mavx512bw -mavx512dq -ldl -lpthread -lm
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.
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 32 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 500M 1000M 1500M 2000M 2500M SE +/- 3481538.61, N = 3 SE +/- 5660683.12, N = 3 SE +/- 1123981.02, N = 3 SE +/- 3540872.46, N = 3 SE +/- 9321659.37, N = 3 SE +/- 4493328.39, N = 3 SE +/- 2804955.93, N = 3 SE +/- 3401960.22, N = 3 2319266667 2308200000 2317400000 2311866667 2268800000 2284300000 2282066667 2281500000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 32 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 400M 800M 1200M 1600M 2000M Min: 2312400000 / Avg: 2319266666.67 / Max: 2323700000 Min: 2297300000 / Avg: 2308200000 / Max: 2316300000 Min: 2315900000 / Avg: 2317400000 / Max: 2319600000 Min: 2305000000 / Avg: 2311866666.67 / Max: 2316800000 Min: 2250400000 / Avg: 2268800000 / Max: 2280600000 Min: 2275600000 / Avg: 2284300000 / Max: 2290600000 Min: 2277700000 / Avg: 2282066666.67 / Max: 2287300000 Min: 2277900000 / Avg: 2281500000 / Max: 2288300000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 64 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1000M 2000M 3000M 4000M 5000M SE +/- 2017699.46, N = 3 SE +/- 7785956.02, N = 3 SE +/- 8247895.35, N = 3 SE +/- 8029321.27, N = 3 SE +/- 3507610.77, N = 3 SE +/- 14090698.27, N = 3 SE +/- 1985782.80, N = 3 SE +/- 7438936.60, N = 3 4349733333 4612933333 4614166667 4625400000 4054500000 4547533333 4564800000 4551866667 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 64 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 800M 1600M 2400M 3200M 4000M Min: 4346800000 / Avg: 4349733333.33 / Max: 4353600000 Min: 4604800000 / Avg: 4612933333.33 / Max: 4628500000 Min: 4600500000 / Avg: 4614166666.67 / Max: 4629000000 Min: 4609500000 / Avg: 4625400000 / Max: 4635300000 Min: 4047500000 / Avg: 4054500000 / Max: 4058400000 Min: 4531300000 / Avg: 4547533333.33 / Max: 4575600000 Min: 4561000000 / Avg: 4564800000 / Max: 4567700000 Min: 4537000000 / Avg: 4551866666.67 / Max: 4559800000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 128 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2000M 4000M 6000M 8000M 10000M SE +/- 1467802.59, N = 3 SE +/- 2027587.51, N = 3 SE +/- 7408178.66, N = 3 SE +/- 11228881.41, N = 3 SE +/- 1556705.92, N = 3 SE +/- 2200000.00, N = 3 SE +/- 3501904.24, N = 3 SE +/- 9653036.37, N = 3 4435766667 4854166667 8545966667 9226733333 5174800000 6182900000 8048400000 9107033333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 128 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1600M 3200M 4800M 6400M 8000M Min: 4434200000 / Avg: 4435766666.67 / Max: 4438700000 Min: 4850500000 / Avg: 4854166666.67 / Max: 4857500000 Min: 8531200000 / Avg: 8545966666.67 / Max: 8554400000 Min: 9213800000 / Avg: 9226733333.33 / Max: 9249100000 Min: 5173000000 / Avg: 5174800000 / Max: 5177900000 Min: 6178500000 / Avg: 6182900000 / Max: 6185100000 Min: 8042600000 / Avg: 8048400000 / Max: 8054700000 Min: 9087800000 / Avg: 9107033333.33 / Max: 9118100000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
Result
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 256 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3000M 6000M 9000M 12000M 15000M SE +/- 8467191.58, N = 3 SE +/- 3002406.44, N = 3 SE +/- 2940710.42, N = 3 SE +/- 3352776.63, N = 3 SE +/- 6811836.10, N = 3 SE +/- 9056182.66, N = 3 SE +/- 4841946.35, N = 3 SE +/- 11259563.84, N = 3 4512600000 4948166667 8713666667 9696366667 5193466667 6140266667 10334333333 12204333333 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid Result Confidence
OpenBenchmarking.org samples/s, More Is Better Liquid-DSP 2021.01.31 Threads: 256 - Buffer Length: 256 - Filter Length: 57 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2000M 4000M 6000M 8000M 10000M Min: 4497200000 / Avg: 4512600000 / Max: 4526400000 Min: 4942400000 / Avg: 4948166666.67 / Max: 4952500000 Min: 8709000000 / Avg: 8713666666.67 / Max: 8719100000 Min: 9690400000 / Avg: 9696366666.67 / Max: 9702000000 Min: 5181100000 / Avg: 5193466666.67 / Max: 5204600000 Min: 6129100000 / Avg: 6140266666.67 / Max: 6158200000 Min: 10329000000 / Avg: 10334333333.33 / Max: 10344000000 Min: 12182000000 / Avg: 12204333333.33 / Max: 12218000000 1. (CC) gcc options: -O3 -pthread -lm -lc -lliquid
PJSIP PJSIP is a free and open source multimedia communication library written in C language implementing standard based protocols such as SIP, SDP, RTP, STUN, TURN, and ICE. It combines signaling protocol (SIP) with rich multimedia framework and NAT traversal functionality into high level API that is portable and suitable for almost any type of systems ranging from desktops, embedded systems, to mobile handsets. This test profile is making use of pjsip-perf with both the client/server on teh system. More details on the PJSIP benchmark at https://www.pjsip.org/high-performance-sip.htm Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateless EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K SE +/- 306.04, N = 3 SE +/- 685.00, N = 3 SE +/- 839.32, N = 15 SE +/- 758.24, N = 15 SE +/- 278.58, N = 3 SE +/- 260.59, N = 3 SE +/- 299.00, N = 3 SE +/- 871.18, N = 5 89733 89834 83661 84745 87898 87728 86749 83007 1. (CC) gcc options: -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2 Result Confidence
OpenBenchmarking.org Responses Per Second, More Is Better PJSIP 2.11 Method: OPTIONS, Stateless EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 16K 32K 48K 64K 80K Min: 89137 / Avg: 89733.33 / Max: 90151 Min: 88574 / Avg: 89833.67 / Max: 90930 Min: 76134 / Avg: 83661.4 / Max: 87258 Min: 78235 / Avg: 84745 / Max: 88561 Min: 87360 / Avg: 87897.67 / Max: 88293 Min: 87241 / Avg: 87728.33 / Max: 88132 Min: 86243 / Avg: 86749 / Max: 87278 Min: 81036 / Avg: 83007 / Max: 85707 1. (CC) gcc options: -lavformat -lavcodec -lswscale -lavutil -lstdc++ -lopus -lssl -lcrypto -luuid -lm -lrt -lpthread -lasound -O2
QuantLib QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.21 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 700 1400 2100 2800 3500 SE +/- 31.02, N = 3 SE +/- 31.45, N = 3 SE +/- 2.89, N = 3 SE +/- 22.61, N = 3 SE +/- 35.26, N = 3 SE +/- 30.22, N = 3 SE +/- 5.62, N = 3 SE +/- 24.81, N = 9 3075.1 3089.5 3106.4 3059.9 3028.2 3043.0 3049.8 3030.6 1. (CXX) g++ options: -O3 -march=native -rdynamic Result Confidence
OpenBenchmarking.org MFLOPS, More Is Better QuantLib 1.21 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 500 1000 1500 2000 2500 Min: 3013.1 / Avg: 3075.13 / Max: 3106.9 Min: 3026.6 / Avg: 3089.47 / Max: 3122.6 Min: 3100.8 / Avg: 3106.43 / Max: 3110.4 Min: 3017 / Avg: 3059.93 / Max: 3093.7 Min: 2957.8 / Avg: 3028.23 / Max: 3066.3 Min: 2982.6 / Avg: 3043.03 / Max: 3074.4 Min: 3038.6 / Avg: 3049.83 / Max: 3055.6 Min: 2834.2 / Avg: 3030.57 / Max: 3066.5 1. (CXX) g++ options: -O3 -march=native -rdynamic
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.
Result
OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 11K 22K 33K 44K 55K SE +/- 21.02, N = 3 SE +/- 10.36, N = 3 SE +/- 19.62, N = 3 SE +/- 12.09, N = 3 SE +/- 0.35, N = 3 SE +/- 27.29, N = 3 SE +/- 25.12, N = 3 SE +/- 95.15, N = 3 17536.0 17546.1 34900.0 35088.9 22205.3 25901.8 44516.7 51406.6 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl Result Confidence
OpenBenchmarking.org sign/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 9K 18K 27K 36K 45K Min: 17494 / Avg: 17536.03 / Max: 17557.8 Min: 17525.4 / Avg: 17546.1 / Max: 17557.1 Min: 34867 / Avg: 34900 / Max: 34934.9 Min: 35064.7 / Avg: 35088.87 / Max: 35101.4 Min: 22204.7 / Avg: 22205.33 / Max: 22205.9 Min: 25867.3 / Avg: 25901.83 / Max: 25955.7 Min: 44466.5 / Avg: 44516.73 / Max: 44542.3 Min: 51242.7 / Avg: 51406.57 / Max: 51572.3 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Result
OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 700K 1400K 2100K 2800K 3500K SE +/- 79.86, N = 3 SE +/- 60.43, N = 3 SE +/- 160.33, N = 3 SE +/- 169.85, N = 3 SE +/- 93.24, N = 3 SE +/- 137.73, N = 3 SE +/- 457.50, N = 3 SE +/- 708.03, N = 3 1141402.4 1141575.6 2282702.1 2282801.7 1463630.7 1689673.1 2936111.9 3369633.3 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl Result Confidence
OpenBenchmarking.org verify/s, More Is Better OpenSSL 3.0 Algorithm: RSA4096 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 600K 1200K 1800K 2400K 3000K Min: 1141307.5 / Avg: 1141402.4 / Max: 1141561.1 Min: 1141456.1 / Avg: 1141575.57 / Max: 1141651.1 Min: 2282386.6 / Avg: 2282702.13 / Max: 2282909.4 Min: 2282485.7 / Avg: 2282801.67 / Max: 2283067.7 Min: 1463447.2 / Avg: 1463630.73 / Max: 1463751.1 Min: 1689402.5 / Avg: 1689673.13 / Max: 1689852.9 Min: 2935275.2 / Avg: 2936111.9 / Max: 2936851 Min: 3368353 / Avg: 3369633.3 / Max: 3370797.4 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Result
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.0 Algorithm: SHA256 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 60000M 120000M 180000M 240000M 300000M SE +/- 88674415.22, N = 3 SE +/- 113242158.65, N = 3 SE +/- 145742816.61, N = 3 SE +/- 540081125.64, N = 3 SE +/- 380252901.09, N = 3 SE +/- 91643095.12, N = 3 SE +/- 584082922.27, N = 3 SE +/- 281531389.95, N = 3 103690964737 103745920213 205408037813 205338727447 125979444607 143163470057 251511273853 284499818607 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl Result Confidence
OpenBenchmarking.org byte/s, More Is Better OpenSSL 3.0 Algorithm: SHA256 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 50000M 100000M 150000M 200000M 250000M Min: 103519943190 / Avg: 103690964736.67 / Max: 103817134810 Min: 103519691010 / Avg: 103745920213.33 / Max: 103868341810 Min: 205230946320 / Avg: 205408037813.33 / Max: 205697087570 Min: 204788321540 / Avg: 205338727446.67 / Max: 206418823480 Min: 125219623880 / Avg: 125979444606.67 / Max: 126387304100 Min: 143004015960 / Avg: 143163470056.67 / Max: 143321466920 Min: 250387877640 / Avg: 251511273853.33 / Max: 252350360300 Min: 283939324380 / Avg: 284499818606.67 / Max: 284826589230 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Xmrig Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Monero - Hash Count: 1M EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K SE +/- 130.07, N = 3 SE +/- 136.28, N = 3 SE +/- 112.63, N = 4 SE +/- 97.02, N = 4 SE +/- 55.16, N = 3 SE +/- 124.44, N = 4 SE +/- 111.70, N = 5 SE +/- 454.82, N = 5 48513.8 48786.0 81468.5 82034.8 58736.1 63045.5 103903.1 111088.8 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc Result Confidence
OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Monero - Hash Count: 1M EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K Min: 48369.9 / Avg: 48513.77 / Max: 48773.4 Min: 48616.9 / Avg: 48786.03 / Max: 49055.7 Min: 81208.4 / Avg: 81468.53 / Max: 81679.3 Min: 81752.8 / Avg: 82034.8 / Max: 82196.3 Min: 58647.6 / Avg: 58736.13 / Max: 58837.4 Min: 62790.4 / Avg: 63045.5 / Max: 63383.4 Min: 103659.2 / Avg: 103903.06 / Max: 104286.2 Min: 110326.6 / Avg: 111088.84 / Max: 112866.8 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Result
OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Wownero - Hash Count: 1M EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 30K 60K 90K 120K 150K SE +/- 86.61, N = 3 SE +/- 46.14, N = 3 SE +/- 83.46, N = 4 SE +/- 164.95, N = 4 SE +/- 23.53, N = 4 SE +/- 49.38, N = 4 SE +/- 196.84, N = 5 SE +/- 156.60, N = 5 60490.2 61774.2 97269.4 98131.4 75830.9 82117.1 126187.4 137480.4 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc Result Confidence
OpenBenchmarking.org H/s, More Is Better Xmrig 6.12.1 Variant: Wownero - Hash Count: 1M EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 20K 40K 60K 80K 100K Min: 60339.1 / Avg: 60490.2 / Max: 60639.1 Min: 61686.5 / Avg: 61774.2 / Max: 61842.9 Min: 97096.8 / Avg: 97269.4 / Max: 97484.9 Min: 97828.2 / Avg: 98131.43 / Max: 98580.4 Min: 75780.5 / Avg: 75830.85 / Max: 75889.8 Min: 81980.7 / Avg: 82117.08 / Max: 82209.8 Min: 125770.3 / Avg: 126187.36 / Max: 126710.6 Min: 136986.3 / Avg: 137480.44 / Max: 137912 1. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc
Sysbench This is a benchmark of Sysbench with the built-in CPU and memory sub-tests. Sysbench is a scriptable multi-threaded benchmark tool based on LuaJIT. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 1.0.20 Test: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 200K 400K 600K 800K 1000K SE +/- 24.46, N = 3 SE +/- 51.83, N = 3 SE +/- 284.53, N = 3 SE +/- 425.17, N = 3 SE +/- 1162.56, N = 3 SE +/- 28.32, N = 3 SE +/- 1204.93, N = 3 SE +/- 1525.81, N = 3 309133.88 309102.11 615405.64 615220.30 446395.22 457526.55 896449.81 910333.64 1. (CC) gcc options: -O2 -funroll-loops -rdynamic -ldl -laio -lm Result Confidence
OpenBenchmarking.org Events Per Second, More Is Better Sysbench 1.0.20 Test: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 160K 320K 480K 640K 800K Min: 309091.81 / Avg: 309133.88 / Max: 309176.53 Min: 309015.47 / Avg: 309102.11 / Max: 309194.72 Min: 614836.65 / Avg: 615405.64 / Max: 615698.18 Min: 614758.83 / Avg: 615220.3 / Max: 616069.58 Min: 445162.08 / Avg: 446395.22 / Max: 448718.88 Min: 457481.46 / Avg: 457526.55 / Max: 457578.77 Min: 894104.2 / Avg: 896449.81 / Max: 898101.32 Min: 907696.65 / Avg: 910333.64 / Max: 912982.19 1. (CC) gcc options: -O2 -funroll-loops -rdynamic -ldl -laio -lm
m-queens A solver for the N-queens problem with multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 SE +/- 0.036, N = 5 SE +/- 0.024, N = 5 SE +/- 0.028, N = 7 SE +/- 0.026, N = 7 SE +/- 0.033, N = 6 SE +/- 0.014, N = 6 SE +/- 0.023, N = 8 SE +/- 0.014, N = 9 9.619 9.566 5.016 5.029 7.988 6.953 4.217 3.738 1. (CXX) g++ options: -fopenmp -O2 -march=native Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better m-queens 1.2 Time To Solve EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 9.51 / Avg: 9.62 / Max: 9.73 Min: 9.48 / Avg: 9.57 / Max: 9.63 Min: 4.91 / Avg: 5.02 / Max: 5.16 Min: 4.97 / Avg: 5.03 / Max: 5.16 Min: 7.89 / Avg: 7.99 / Max: 8.09 Min: 6.91 / Avg: 6.95 / Max: 7 Min: 4.12 / Avg: 4.22 / Max: 4.33 Min: 3.65 / Avg: 3.74 / Max: 3.8 1. (CXX) g++ options: -fopenmp -O2 -march=native
Coremark This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2M 4M 6M 8M 10M SE +/- 16091.05, N = 3 SE +/- 13633.41, N = 3 SE +/- 6330.65, N = 3 SE +/- 9387.28, N = 3 SE +/- 4348.78, N = 3 SE +/- 17803.94, N = 3 SE +/- 75759.77, N = 3 SE +/- 58302.36, N = 3 2950220.28 2930510.79 6009403.02 6023637.43 3753920.39 4111551.58 7536121.01 8413564.08 1. (CC) gcc options: -O2 -lrt" -lrt Result Confidence
OpenBenchmarking.org Iterations/Sec, More Is Better Coremark 1.0 CoreMark Size 666 - Iterations Per Second EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1.5M 3M 4.5M 6M 7.5M Min: 2924822.91 / Avg: 2950220.28 / Max: 2980036.09 Min: 2910413.82 / Avg: 2930510.79 / Max: 2956518.39 Min: 5996954.67 / Avg: 6009403.02 / Max: 6017629.77 Min: 6006099.95 / Avg: 6023637.43 / Max: 6038210.55 Min: 3747316.37 / Avg: 3753920.39 / Max: 3762124.03 Min: 4088150.75 / Avg: 4111551.58 / Max: 4146495.1 Min: 7384852.08 / Avg: 7536121.01 / Max: 7619299.58 Min: 8297320.66 / Avg: 8413564.08 / Max: 8479629.02 1. (CC) gcc options: -O2 -lrt" -lrt
asmFish This is a test of asmFish, an advanced chess benchmark written in Assembly. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 90M 180M 270M 360M 450M SE +/- 1373257.95, N = 3 SE +/- 835429.68, N = 3 SE +/- 1040134.87, N = 3 SE +/- 3711493.29, N = 5 SE +/- 2828612.84, N = 3 SE +/- 1169064.32, N = 3 SE +/- 2695071.60, N = 3 SE +/- 1520548.58, N = 3 209533141 212598823 368462801 373466857 249175879 280313823 368919863 397046539 Result Confidence
OpenBenchmarking.org Nodes/second, More Is Better asmFish 2018-07-23 1024 Hash Memory, 26 Depth EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 70M 140M 210M 280M 350M Min: 207533584 / Avg: 209533141.33 / Max: 212163515 Min: 210965402 / Avg: 212598823.33 / Max: 213720131 Min: 366401691 / Avg: 368462800.67 / Max: 369737306 Min: 367390642 / Avg: 373466857 / Max: 387387799 Min: 244432430 / Avg: 249175878.67 / Max: 254217420 Min: 278823104 / Avg: 280313823 / Max: 282619135 Min: 365620710 / Avg: 368919863 / Max: 374260906 Min: 394008334 / Avg: 397046539.33 / Max: 398680469
Tachyon This is a test of the threaded Tachyon, a parallel ray-tracing system, measuring the time to ray-trace a sample scene. The sample scene used is the Teapot scene ray-traced to 8K x 8K with 32 samples. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.99.2 Total Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 SE +/- 0.0168, N = 3 SE +/- 0.0045, N = 3 SE +/- 0.0786, N = 5 SE +/- 0.0480, N = 5 SE +/- 0.0101, N = 3 SE +/- 0.0145, N = 4 SE +/- 0.0823, N = 7 SE +/- 0.0339, N = 6 19.9688 18.8063 10.4833 9.8306 16.7586 14.6145 8.8633 7.9829 1. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread Result Confidence
OpenBenchmarking.org Seconds, Fewer Is Better Tachyon 0.99.2 Total Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 Min: 19.94 / Avg: 19.97 / Max: 20 Min: 18.8 / Avg: 18.81 / Max: 18.82 Min: 10.27 / Avg: 10.48 / Max: 10.71 Min: 9.73 / Avg: 9.83 / Max: 10 Min: 16.74 / Avg: 16.76 / Max: 16.78 Min: 14.57 / Avg: 14.61 / Max: 14.64 Min: 8.66 / Avg: 8.86 / Max: 9.25 Min: 7.86 / Avg: 7.98 / Max: 8.09 1. (CC) gcc options: -m64 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread
WebP Image Encode This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Default EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 SE +/- 0.07, N = 15 SE +/- 0.01, N = 12 SE +/- 0.01, N = 12 SE +/- 0.02, N = 15 SE +/- 0.04, N = 15 SE +/- 0.01, N = 12 SE +/- 0.01, N = 12 SE +/- 0.01, N = 12 18.77 18.85 18.85 18.81 18.53 18.59 18.59 18.56 1. (CC) gcc options: -fvisibility=hidden -O2 -lm Result Confidence
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Default EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 5 10 15 20 25 Min: 17.76 / Avg: 18.77 / Max: 18.9 Min: 18.79 / Avg: 18.85 / Max: 18.88 Min: 18.81 / Avg: 18.85 / Max: 18.88 Min: 18.59 / Avg: 18.81 / Max: 18.85 Min: 17.95 / Avg: 18.53 / Max: 18.62 Min: 18.56 / Avg: 18.59 / Max: 18.62 Min: 18.55 / Avg: 18.59 / Max: 18.62 Min: 18.49 / Avg: 18.56 / Max: 18.6 1. (CC) gcc options: -fvisibility=hidden -O2 -lm
Result
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 SE +/- 0.01, N = 10 SE +/- 0.03, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.05, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 SE +/- 0.01, N = 10 11.70 11.69 11.69 11.67 11.48 11.55 11.53 11.52 1. (CC) gcc options: -fvisibility=hidden -O2 -lm Result Confidence
OpenBenchmarking.org MP/s, More Is Better WebP Image Encode 1.2.4 Encode Settings: Quality 100 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 Min: 11.68 / Avg: 11.7 / Max: 11.73 Min: 11.4 / Avg: 11.69 / Max: 11.74 Min: 11.64 / Avg: 11.69 / Max: 11.72 Min: 11.56 / Avg: 11.67 / Max: 11.72 Min: 11.07 / Avg: 11.48 / Max: 11.58 Min: 11.51 / Avg: 11.55 / Max: 11.58 Min: 11.47 / Avg: 11.53 / Max: 11.56 Min: 11.49 / Avg: 11.52 / Max: 11.56 1. (CC) gcc options: -fvisibility=hidden -O2 -lm
WebP2 Image Encode This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Default EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 3 6 9 12 15 SE +/- 0.19, N = 15 SE +/- 0.17, N = 15 SE +/- 0.21, N = 15 SE +/- 0.20, N = 15 SE +/- 0.10, N = 15 SE +/- 0.18, N = 15 SE +/- 0.15, N = 15 SE +/- 0.20, N = 15 11.76 11.69 10.34 10.15 11.60 12.06 9.96 10.21 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl Result Confidence
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Default EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 4 8 12 16 20 Min: 11.11 / Avg: 11.76 / Max: 13.07 Min: 10.97 / Avg: 11.69 / Max: 12.93 Min: 9.47 / Avg: 10.34 / Max: 11.88 Min: 9.59 / Avg: 10.15 / Max: 12.13 Min: 11.37 / Avg: 11.6 / Max: 12.95 Min: 11.47 / Avg: 12.06 / Max: 13.45 Min: 9.24 / Avg: 9.96 / Max: 11.76 Min: 9.06 / Avg: 10.21 / Max: 11.64 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.1913 0.3826 0.5739 0.7652 0.9565 SE +/- 0.00, N = 13 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.72 0.72 0.85 0.84 0.78 0.83 0.83 0.83 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl Result Confidence
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 75, Compression Effort 7 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 0.69 / Avg: 0.72 / Max: 0.76 Min: 0.71 / Avg: 0.72 / Max: 0.73 Min: 0.84 / Avg: 0.85 / Max: 0.85 Min: 0.84 / Avg: 0.84 / Max: 0.84 Min: 0.78 / Avg: 0.78 / Max: 0.78 Min: 0.83 / Avg: 0.83 / Max: 0.84 Min: 0.83 / Avg: 0.83 / Max: 0.83 Min: 0.82 / Avg: 0.83 / Max: 0.83 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
Result
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.1035 0.207 0.3105 0.414 0.5175 SE +/- 0.00, N = 3 SE +/- 0.00, N = 4 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.36 0.36 0.46 0.46 0.41 0.45 0.46 0.45 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl Result Confidence
OpenBenchmarking.org MP/s, More Is Better WebP2 Image Encode 20220823 Encode Settings: Quality 95, Compression Effort 7 EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1 2 3 4 5 Min: 0.36 / Avg: 0.36 / Max: 0.36 Min: 0.35 / Avg: 0.36 / Max: 0.38 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.46 / Avg: 0.46 / Max: 0.46 Min: 0.41 / Avg: 0.41 / Max: 0.41 Min: 0.45 / Avg: 0.45 / Max: 0.45 Min: 0.45 / Avg: 0.46 / Max: 0.46 Min: 0.45 / Avg: 0.45 / Max: 0.45 1. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl
OpenFOAM OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Mesh Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P 7 14 21 28 35 22.84 22.85 21.93 22.29 25.94 23.19 31.67 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Small Mesh Size - Execution Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P 7 14 21 28 35 31.21 30.32 20.98 21.20 28.52 25.14 22.88 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Mesh Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P 30 60 90 120 150 102.78 110.84 109.42 106.44 131.29 120.42 129.44 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
OpenBenchmarking.org Seconds, Fewer Is Better OpenFOAM 10 Input: drivaerFastback, Medium Mesh Size - Execution Time EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P 60 120 180 240 300 297.87 298.39 129.81 129.17 246.66 237.94 111.57 1. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm
oneDNN Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.1507 0.3014 0.4521 0.6028 0.7535 SE +/- 0.001014, N = 7 SE +/- 0.000630, N = 7 SE +/- 0.002491, N = 7 SE +/- 0.003117, N = 7 SE +/- 0.000897, N = 7 SE +/- 0.001009, N = 7 SE +/- 0.003944, N = 15 SE +/- 0.003698, N = 15 0.597698 0.585913 0.439404 0.438693 0.669803 0.610119 0.529838 0.488721 MIN: 0.56 MIN: 0.57 MIN: 0.37 MIN: 0.36 MIN: 0.6 MIN: 0.58 MIN: 0.44 MIN: 0.41 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 2 4 6 8 10 Min: 0.59 / Avg: 0.6 / Max: 0.6 Min: 0.58 / Avg: 0.59 / Max: 0.59 Min: 0.43 / Avg: 0.44 / Max: 0.45 Min: 0.43 / Avg: 0.44 / Max: 0.45 Min: 0.67 / Avg: 0.67 / Max: 0.67 Min: 0.61 / Avg: 0.61 / Max: 0.62 Min: 0.5 / Avg: 0.53 / Max: 0.55 Min: 0.47 / Avg: 0.49 / Max: 0.51 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.086 0.172 0.258 0.344 0.43 SE +/- 0.000785, N = 7 SE +/- 0.000585, N = 7 SE +/- 0.001079, N = 7 SE +/- 0.002431, N = 15 SE +/- 0.001445, N = 7 SE +/- 0.001337, N = 7 SE +/- 0.002791, N = 7 SE +/- 0.003853, N = 15 0.281845 0.260069 0.294229 0.284125 0.290728 0.275194 0.382159 0.351182 MIN: 0.25 MIN: 0.24 MIN: 0.22 MIN: 0.21 MIN: 0.27 MIN: 0.25 MIN: 0.28 MIN: 0.27 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1 2 3 4 5 Min: 0.28 / Avg: 0.28 / Max: 0.28 Min: 0.26 / Avg: 0.26 / Max: 0.26 Min: 0.29 / Avg: 0.29 / Max: 0.3 Min: 0.26 / Avg: 0.28 / Max: 0.3 Min: 0.29 / Avg: 0.29 / Max: 0.3 Min: 0.27 / Avg: 0.28 / Max: 0.28 Min: 0.37 / Avg: 0.38 / Max: 0.4 Min: 0.33 / Avg: 0.35 / Max: 0.38 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
Result
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 0.0994 0.1988 0.2982 0.3976 0.497 SE +/- 0.001053, N = 7 SE +/- 0.000790, N = 7 SE +/- 0.002017, N = 7 SE +/- 0.002384, N = 15 SE +/- 0.001543, N = 7 SE +/- 0.001518, N = 7 SE +/- 0.001212, N = 7 SE +/- 0.003312, N = 7 0.400531 0.397186 0.359965 0.352990 0.394456 0.375184 0.441567 0.386561 MIN: 0.38 MIN: 0.38 MIN: 0.29 MIN: 0.29 MIN: 0.37 MIN: 0.35 MIN: 0.34 MIN: 0.32 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread Result Confidence
OpenBenchmarking.org ms, Fewer Is Better oneDNN 2.7 Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPU EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power EPYC 9654 2P EPYC 9654 2P - Power 1 2 3 4 5 Min: 0.4 / Avg: 0.4 / Max: 0.41 Min: 0.4 / Avg: 0.4 / Max: 0.4 Min: 0.36 / Avg: 0.36 / Max: 0.37 Min: 0.33 / Avg: 0.35 / Max: 0.36 Min: 0.39 / Avg: 0.39 / Max: 0.4 Min: 0.37 / Avg: 0.38 / Max: 0.38 Min: 0.44 / Avg: 0.44 / Max: 0.45 Min: 0.38 / Avg: 0.39 / Max: 0.41 1. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl -lpthread
Neural Magic DeepSparse Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 SE +/- 0.19, N = 3 SE +/- 0.12, N = 3 SE +/- 0.23, N = 3 SE +/- 0.16, N = 3 SE +/- 0.21, N = 3 SE +/- 0.11, N = 3 89.03 89.10 88.20 88.05 87.52 89.50 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 Min: 88.67 / Avg: 89.03 / Max: 89.29 Min: 88.88 / Avg: 89.1 / Max: 89.31 Min: 87.74 / Avg: 88.2 / Max: 88.5 Min: 87.85 / Avg: 88.05 / Max: 88.37 Min: 87.2 / Avg: 87.52 / Max: 87.91 Min: 89.29 / Avg: 89.5 / Max: 89.62
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 11.23 11.22 11.33 11.35 11.42 11.17 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 3 6 9 12 15 Min: 11.19 / Avg: 11.23 / Max: 11.27 Min: 11.19 / Avg: 11.22 / Max: 11.25 Min: 11.29 / Avg: 11.33 / Max: 11.39 Min: 11.31 / Avg: 11.35 / Max: 11.38 Min: 11.37 / Avg: 11.42 / Max: 11.46 Min: 11.15 / Avg: 11.17 / Max: 11.19
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 120 240 360 480 600 SE +/- 0.57, N = 3 SE +/- 0.48, N = 3 SE +/- 0.61, N = 3 SE +/- 2.06, N = 3 SE +/- 1.28, N = 3 SE +/- 1.13, N = 3 262.96 281.83 509.13 557.55 316.56 336.60 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 100 200 300 400 500 Min: 261.82 / Avg: 262.96 / Max: 263.6 Min: 280.89 / Avg: 281.83 / Max: 282.4 Min: 507.97 / Avg: 509.13 / Max: 510.04 Min: 554.7 / Avg: 557.55 / Max: 561.55 Min: 314.05 / Avg: 316.56 / Max: 318.26 Min: 334.48 / Avg: 336.6 / Max: 338.33
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 30 60 90 120 150 SE +/- 0.23, N = 3 SE +/- 0.18, N = 3 SE +/- 0.15, N = 3 SE +/- 0.41, N = 3 SE +/- 0.55, N = 3 SE +/- 0.50, N = 3 121.43 113.41 125.42 114.56 151.40 142.33 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 30 60 90 120 150 Min: 121.17 / Avg: 121.43 / Max: 121.9 Min: 113.16 / Avg: 113.41 / Max: 113.77 Min: 125.19 / Avg: 125.42 / Max: 125.7 Min: 113.74 / Avg: 114.56 / Max: 115.01 Min: 150.67 / Avg: 151.4 / Max: 152.47 Min: 141.54 / Avg: 142.33 / Max: 143.24
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 40 80 120 160 200 SE +/- 0.47, N = 3 SE +/- 0.28, N = 3 SE +/- 0.29, N = 3 SE +/- 0.60, N = 3 SE +/- 0.35, N = 3 SE +/- 0.52, N = 3 172.18 172.26 167.07 168.25 189.88 196.20 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 40 80 120 160 200 Min: 171.49 / Avg: 172.18 / Max: 173.09 Min: 171.75 / Avg: 172.26 / Max: 172.72 Min: 166.77 / Avg: 167.07 / Max: 167.66 Min: 167.34 / Avg: 168.25 / Max: 169.4 Min: 189.19 / Avg: 189.88 / Max: 190.31 Min: 195.49 / Avg: 196.2 / Max: 197.2
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 1.3451 2.6902 4.0353 5.3804 6.7255 SE +/- 0.0160, N = 3 SE +/- 0.0095, N = 3 SE +/- 0.0103, N = 3 SE +/- 0.0211, N = 3 SE +/- 0.0098, N = 3 SE +/- 0.0133, N = 3 5.8020 5.7994 5.9782 5.9369 5.2631 5.0933 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 2 4 6 8 10 Min: 5.77 / Avg: 5.8 / Max: 5.83 Min: 5.78 / Avg: 5.8 / Max: 5.82 Min: 5.96 / Avg: 5.98 / Max: 5.99 Min: 5.9 / Avg: 5.94 / Max: 5.97 Min: 5.25 / Avg: 5.26 / Max: 5.28 Min: 5.07 / Avg: 5.09 / Max: 5.11
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 200 400 600 800 1000 SE +/- 0.71, N = 3 SE +/- 1.32, N = 3 SE +/- 0.91, N = 3 SE +/- 3.40, N = 3 SE +/- 2.07, N = 3 SE +/- 1.04, N = 3 520.75 557.26 996.36 1090.99 626.04 666.25 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 200 400 600 800 1000 Min: 519.6 / Avg: 520.75 / Max: 522.06 Min: 555.13 / Avg: 557.26 / Max: 559.67 Min: 995.23 / Avg: 996.36 / Max: 998.15 Min: 1084.5 / Avg: 1090.99 / Max: 1096.02 Min: 622.68 / Avg: 626.04 / Max: 629.81 Min: 664.17 / Avg: 666.25 / Max: 667.45
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 SE +/- 0.08, N = 3 SE +/- 0.14, N = 3 SE +/- 0.06, N = 3 SE +/- 0.17, N = 3 SE +/- 0.26, N = 3 SE +/- 0.11, N = 3 61.38 57.40 64.12 58.53 76.60 71.96 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 15 30 45 60 75 Min: 61.24 / Avg: 61.38 / Max: 61.53 Min: 57.15 / Avg: 57.4 / Max: 57.62 Min: 64.02 / Avg: 64.12 / Max: 64.21 Min: 58.29 / Avg: 58.53 / Max: 58.86 Min: 76.12 / Avg: 76.6 / Max: 77.01 Min: 71.81 / Avg: 71.96 / Max: 72.18
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 60 120 180 240 300 SE +/- 0.30, N = 3 SE +/- 0.76, N = 3 SE +/- 0.47, N = 3 SE +/- 0.50, N = 3 SE +/- 0.43, N = 3 SE +/- 0.23, N = 3 252.55 255.83 256.65 255.74 192.24 192.52 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 50 100 150 200 250 Min: 252.12 / Avg: 252.55 / Max: 253.12 Min: 254.65 / Avg: 255.83 / Max: 257.26 Min: 255.89 / Avg: 256.65 / Max: 257.5 Min: 255.08 / Avg: 255.74 / Max: 256.73 Min: 191.57 / Avg: 192.24 / Max: 193.03 Min: 192.24 / Avg: 192.52 / Max: 192.98
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 1.17 2.34 3.51 4.68 5.85 SE +/- 0.0046, N = 3 SE +/- 0.0116, N = 3 SE +/- 0.0072, N = 3 SE +/- 0.0077, N = 3 SE +/- 0.0115, N = 3 SE +/- 0.0062, N = 3 3.9577 3.9067 3.8941 3.9080 5.1998 5.1923 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 2 4 6 8 10 Min: 3.95 / Avg: 3.96 / Max: 3.96 Min: 3.89 / Avg: 3.91 / Max: 3.92 Min: 3.88 / Avg: 3.89 / Max: 3.91 Min: 3.89 / Avg: 3.91 / Max: 3.92 Min: 5.18 / Avg: 5.2 / Max: 5.22 Min: 5.18 / Avg: 5.19 / Max: 5.2
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 400 800 1200 1600 2000 SE +/- 0.93, N = 3 SE +/- 0.74, N = 3 SE +/- 0.15, N = 3 SE +/- 1.89, N = 3 SE +/- 0.42, N = 3 SE +/- 0.46, N = 3 843.37 879.22 1615.51 1739.26 1014.62 1133.41 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 300 600 900 1200 1500 Min: 842.04 / Avg: 843.37 / Max: 845.15 Min: 877.79 / Avg: 879.22 / Max: 880.24 Min: 1615.3 / Avg: 1615.51 / Max: 1615.79 Min: 1735.76 / Avg: 1739.26 / Max: 1742.26 Min: 1014.11 / Avg: 1014.62 / Max: 1015.45 Min: 1132.5 / Avg: 1133.41 / Max: 1133.9
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 11 22 33 44 55 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.00, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 37.89 36.35 39.55 36.73 47.25 42.29 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 10 20 30 40 50 Min: 37.8 / Avg: 37.89 / Max: 37.95 Min: 36.3 / Avg: 36.35 / Max: 36.41 Min: 39.55 / Avg: 39.55 / Max: 39.55 Min: 36.65 / Avg: 36.73 / Max: 36.82 Min: 47.22 / Avg: 47.25 / Max: 47.27 Min: 42.24 / Avg: 42.29 / Max: 42.32
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 9 18 27 36 45 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.07, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.08, N = 3 26.44 26.58 26.11 26.39 35.57 37.25 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 8 16 24 32 40 Min: 26.42 / Avg: 26.44 / Max: 26.47 Min: 26.56 / Avg: 26.58 / Max: 26.61 Min: 26.01 / Avg: 26.11 / Max: 26.24 Min: 26.25 / Avg: 26.39 / Max: 26.48 Min: 35.51 / Avg: 35.57 / Max: 35.69 Min: 37.15 / Avg: 37.25 / Max: 37.4
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 9 18 27 36 45 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 37.82 37.62 38.29 37.88 28.11 26.84 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 8 16 24 32 40 Min: 37.78 / Avg: 37.82 / Max: 37.84 Min: 37.57 / Avg: 37.62 / Max: 37.65 Min: 38.1 / Avg: 38.29 / Max: 38.45 Min: 37.76 / Avg: 37.88 / Max: 38.08 Min: 28.01 / Avg: 28.11 / Max: 28.16 Min: 26.73 / Avg: 26.84 / Max: 26.92
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 16 32 48 64 80 SE +/- 0.06, N = 3 SE +/- 0.02, N = 3 SE +/- 0.15, N = 3 SE +/- 0.05, N = 3 SE +/- 0.31, N = 3 SE +/- 0.03, N = 3 35.66 36.98 70.05 73.51 42.69 49.10 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 14 28 42 56 70 Min: 35.54 / Avg: 35.66 / Max: 35.74 Min: 36.96 / Avg: 36.98 / Max: 37.01 Min: 69.74 / Avg: 70.05 / Max: 70.23 Min: 73.46 / Avg: 73.51 / Max: 73.6 Min: 42.1 / Avg: 42.69 / Max: 43.14 Min: 49.05 / Avg: 49.1 / Max: 49.14
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 SE +/- 0.38, N = 3 SE +/- 0.12, N = 3 SE +/- 0.43, N = 3 SE +/- 0.94, N = 3 SE +/- 0.29, N = 3 SE +/- 0.13, N = 3 92.68 93.34 91.50 90.94 101.65 102.65 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 Min: 91.93 / Avg: 92.68 / Max: 93.11 Min: 93.11 / Avg: 93.34 / Max: 93.53 Min: 90.95 / Avg: 91.5 / Max: 92.35 Min: 89.15 / Avg: 90.94 / Max: 92.33 Min: 101.08 / Avg: 101.65 / Max: 102 Min: 102.47 / Avg: 102.65 / Max: 102.9
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 3 6 9 12 15 SE +/- 0.0441, N = 3 SE +/- 0.0141, N = 3 SE +/- 0.0509, N = 3 SE +/- 0.1144, N = 3 SE +/- 0.0277, N = 3 SE +/- 0.0122, N = 3 10.7833 10.7061 10.9209 10.9903 9.8328 9.7375 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 3 6 9 12 15 Min: 10.73 / Avg: 10.78 / Max: 10.87 Min: 10.69 / Avg: 10.71 / Max: 10.73 Min: 10.82 / Avg: 10.92 / Max: 10.99 Min: 10.82 / Avg: 10.99 / Max: 11.21 Min: 9.8 / Avg: 9.83 / Max: 9.89 Min: 9.71 / Avg: 9.74 / Max: 9.75
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 150 300 450 600 750 SE +/- 0.26, N = 3 SE +/- 0.30, N = 3 SE +/- 0.32, N = 3 SE +/- 0.55, N = 3 SE +/- 0.33, N = 3 SE +/- 0.50, N = 3 329.12 358.43 629.11 694.62 401.20 428.44 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 120 240 360 480 600 Min: 328.6 / Avg: 329.12 / Max: 329.41 Min: 357.99 / Avg: 358.43 / Max: 359 Min: 628.64 / Avg: 629.11 / Max: 629.73 Min: 693.52 / Avg: 694.62 / Max: 695.25 Min: 400.69 / Avg: 401.2 / Max: 401.8 Min: 427.52 / Avg: 428.44 / Max: 429.26
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 30 60 90 120 150 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 SE +/- 0.10, N = 3 SE +/- 0.15, N = 3 97.09 89.13 101.51 91.94 119.41 111.75 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 Min: 97.01 / Avg: 97.09 / Max: 97.26 Min: 88.97 / Avg: 89.13 / Max: 89.27 Min: 101.41 / Avg: 101.51 / Max: 101.57 Min: 91.87 / Avg: 91.94 / Max: 92.06 Min: 119.23 / Avg: 119.41 / Max: 119.57 Min: 111.54 / Avg: 111.75 / Max: 112.03
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 40 80 120 160 200 SE +/- 0.08, N = 3 SE +/- 0.13, N = 3 SE +/- 0.16, N = 3 SE +/- 0.14, N = 3 SE +/- 0.11, N = 3 SE +/- 0.33, N = 3 189.26 189.93 188.04 189.73 192.31 199.58 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 40 80 120 160 200 Min: 189.15 / Avg: 189.26 / Max: 189.42 Min: 189.69 / Avg: 189.93 / Max: 190.13 Min: 187.72 / Avg: 188.04 / Max: 188.21 Min: 189.58 / Avg: 189.73 / Max: 190 Min: 192.19 / Avg: 192.31 / Max: 192.52 Min: 199 / Avg: 199.58 / Max: 200.15
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 1.1955 2.391 3.5865 4.782 5.9775 SE +/- 0.0022, N = 3 SE +/- 0.0035, N = 3 SE +/- 0.0045, N = 3 SE +/- 0.0039, N = 3 SE +/- 0.0029, N = 3 SE +/- 0.0082, N = 3 5.2790 5.2608 5.3134 5.2659 5.1960 5.0062 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 2 4 6 8 10 Min: 5.27 / Avg: 5.28 / Max: 5.28 Min: 5.26 / Avg: 5.26 / Max: 5.27 Min: 5.31 / Avg: 5.31 / Max: 5.32 Min: 5.26 / Avg: 5.27 / Max: 5.27 Min: 5.19 / Avg: 5.2 / Max: 5.2 Min: 4.99 / Avg: 5.01 / Max: 5.02
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 200 400 600 800 1000 SE +/- 0.64, N = 3 SE +/- 0.23, N = 3 SE +/- 1.74, N = 3 SE +/- 2.55, N = 3 SE +/- 0.82, N = 3 SE +/- 0.57, N = 3 364.70 388.32 710.90 776.30 447.55 499.75 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 140 280 420 560 700 Min: 363.81 / Avg: 364.7 / Max: 365.94 Min: 387.9 / Avg: 388.32 / Max: 388.69 Min: 707.59 / Avg: 710.9 / Max: 713.5 Min: 772.32 / Avg: 776.3 / Max: 781.04 Min: 445.98 / Avg: 447.55 / Max: 448.77 Min: 498.84 / Avg: 499.75 / Max: 500.81
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 SE +/- 0.18, N = 3 SE +/- 0.07, N = 3 SE +/- 0.23, N = 3 SE +/- 0.27, N = 3 SE +/- 0.19, N = 3 SE +/- 0.13, N = 3 87.56 82.29 89.81 82.26 107.01 95.85 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 20 40 60 80 100 Min: 87.22 / Avg: 87.56 / Max: 87.81 Min: 82.18 / Avg: 82.29 / Max: 82.42 Min: 89.43 / Avg: 89.81 / Max: 90.23 Min: 81.76 / Avg: 82.26 / Max: 82.7 Min: 106.71 / Avg: 107.01 / Max: 107.35 Min: 95.6 / Avg: 95.85 / Max: 96.02
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 9 18 27 36 45 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.08, N = 3 SE +/- 0.15, N = 3 SE +/- 0.07, N = 3 26.44 26.49 26.14 26.28 35.66 37.15 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 8 16 24 32 40 Min: 26.34 / Avg: 26.44 / Max: 26.59 Min: 26.37 / Avg: 26.49 / Max: 26.64 Min: 26.11 / Avg: 26.14 / Max: 26.18 Min: 26.17 / Avg: 26.28 / Max: 26.43 Min: 35.48 / Avg: 35.66 / Max: 35.95 Min: 37.01 / Avg: 37.15 / Max: 37.24
Result
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 9 18 27 36 45 SE +/- 0.10, N = 3 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 SE +/- 0.11, N = 3 SE +/- 0.11, N = 3 SE +/- 0.05, N = 3 37.81 37.74 38.25 38.04 28.04 26.91 Result Confidence
OpenBenchmarking.org ms/batch, Fewer Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Synchronous Single-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 8 16 24 32 40 Min: 37.61 / Avg: 37.81 / Max: 37.95 Min: 37.54 / Avg: 37.74 / Max: 37.92 Min: 38.19 / Avg: 38.25 / Max: 38.29 Min: 37.83 / Avg: 38.04 / Max: 38.21 Min: 27.81 / Avg: 28.04 / Max: 28.18 Min: 26.84 / Avg: 26.91 / Max: 27.01
Result
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 16 32 48 64 80 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.47, N = 3 SE +/- 0.26, N = 3 SE +/- 0.36, N = 3 SE +/- 0.04, N = 3 35.69 36.96 69.62 73.12 42.52 48.98 Result Confidence
OpenBenchmarking.org items/sec, More Is Better Neural Magic DeepSparse 1.1 Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Stream EPYC 9554 EPYC 9554 - Power EPYC 9554 2P EPYC 9554 2P - Power EPYC 9654 EPYC 9654 - Power 14 28 42 56 70 Min: 35.63 / Avg: 35.69 / Max: 35.76 Min: 36.95 / Avg: 36.96 / Max: 36.98 Min: 68.7 / Avg: 69.62 / Max: 70.27 Min: 72.85 / Avg: 73.12 / Max: 73.64 Min: 42.11 / Avg: 42.52 / Max: 43.24 Min: 48.91 / Avg: 48.98 / Max: 49.05
OpenRadioss OpenRadioss is an open-source AGPL-licensed finite element solver for dynamic event analysis OpenRadioss is based on Altair Radioss and open-sourced in 2022. This open-source finite element solver is benchmarked with various example models available from https://www.openradioss.org/models/. This test is currently using a reference OpenRadioss binary build offered via GitHub. Learn more via the OpenBenchmarking.org test page.
EPYC 9554 Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 2 November 2022 18:00 by user phoronix.
EPYC 9554 - Power Processor: AMD EPYC 9554 64-Core @ 3.76GHz (64 Cores / 128 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 2 November 2022 04:00 by user phoronix.
EPYC 9554 2P Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 31 October 2022 16:00 by user phoronix.
EPYC 9554 2P - Power Processor: 2 x AMD EPYC 9554 64-Core @ 3.76GHz (128 Cores / 256 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 1 November 2022 14:00 by user phoronix.
EPYC 9654 Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 15 October 2022 20:00 by user phoronix.
EPYC 9654 - Power Processor: AMD EPYC 9654 96-Core @ 3.71GHz (96 Cores / 192 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 768GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 16 October 2022 16:00 by user phoronix.
EPYC 9654 2P Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 7 October 2022 01:00 by user phoronix.
EPYC 9654 2P - Power Processor: 2 x AMD EPYC 9654 96-Core @ 3.71GHz (192 Cores / 384 Threads), Motherboard: AMD Titanite_4G (RTI1002E BIOS), Chipset: AMD Device 14a4, Memory: 1520GB, Disk: 3841GB Micron_9300_MTFDHAL3T8TDP, Graphics: ASPEED, Monitor: VGA HDMI, Network: Broadcom NetXtreme BCM5720 PCIe
OS: Ubuntu 22.10, Kernel: 6.0.0-060000rc3daily20220904-generic (x86_64), Desktop: GNOME Shell, Display Server: X Server 1.21.1.3, Vulkan: 1.3.211, Compiler: GCC 12.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-12-Wbc0TK/gcc-12-12.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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: Scaling Governor: amd-pstate performance (Boost: Enabled) - CPU Microcode: 0xa10110dJava Notes: OpenJDK Runtime Environment (build 11.0.16+8-post-Ubuntu-0ubuntu1)Python 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 usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Retpolines IBPB: conditional IBRS_FW STIBP: always-on RSB filling PBRSB-eIBRS: Not affected + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 7 October 2022 21:00 by user phoronix.