AMD Ryzen Threadripper 2990WX compiler benchmarks on GCC 9.1 with Ubuntu Linux by Michael Larabel.
-O2 -march=athlon64 Environment Notes: CXXFLAGS=-O2-march=athlon64 CFLAGS=-O2-march=athlon64Compiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 -march=athlon64 Environment Notes: CXXFLAGS=-O3-march=athlon64 CFLAGS=-O3-march=athlon64Compiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 -march=athlon64-sse3 Environment Notes: CXXFLAGS=-O3-march=athlon64-sse3 CFLAGS=-O3-march=athlon64-sse3Compiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O2 -march=native Environment Notes: CXXFLAGS=-O2-march=native CFLAGS=-O2-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 -march=native Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
-O3 -march=native -flto Environment Notes: CXXFLAGS=-O3-march=native-flto CFLAGS=-O3-march=native-fltoCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
PGO Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
AMD Ryzen Threadripper 2990WX 32-Core Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad
OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160
MBW This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 8192 MiB PGO AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=native -flto 3K 6K 9K 12K 15K SE +/- 61.39, N = 3 SE +/- 152.24, N = 5 SE +/- 217.78, N = 3 SE +/- 167.95, N = 5 SE +/- 169.00, N = 3 SE +/- 162.89, N = 5 SE +/- 153.84, N = 3 12406 12618 12632 12699 12795 12851 12930 13176 -O2 -march=athlon64 -march=athlon64-sse3 -O2 -march=athlon64 -flto 1. (CC) gcc options: -O3 -march=native
OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 8192 MiB -O2 -march=athlon64 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto PGO -O3 -march=athlon64 -O2 -march=native 1500 3000 4500 6000 7500 SE +/- 60.66, N = 3 SE +/- 10.14, N = 3 SE +/- 27.83, N = 3 SE +/- 96.68, N = 3 SE +/- 26.90, N = 3 SE +/- 7.74, N = 3 SE +/- 38.61, N = 3 6526 6596 6681 6685 6721 6753 6755 6778 -O2 -march=athlon64 -march=athlon64-sse3 -flto -march=athlon64 -O2 1. (CC) gcc options: -O3 -march=native
t-test1 This is a test of t-test1 for basic memory allocator benchmarks. Note this test profile is currently very basic and the overall time does include the warmup time of the custom t-test1 compilation. Improvements welcome. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 1 -O3 -march=athlon64 PGO AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64-sse3 -O3 -march=native -flto -O3 -march=native -O2 -march=native -O2 -march=athlon64 7 14 21 28 35 SE +/- 0.29, N = 9 SE +/- 0.39, N = 3 SE +/- 0.34, N = 15 SE +/- 0.16, N = 3 SE +/- 0.07, N = 3 SE +/- 0.33, N = 3 SE +/- 0.10, N = 3 29.66 28.74 27.83 27.12 26.31 26.23 26.07 25.60 -O3 -march=athlon64 -O3 -march=native -O3 -march=native -O3 -march=athlon64-sse3 -O3 -march=native -flto -O3 -march=native -O2 -march=native -O2 -march=athlon64 1. (CC) gcc options: -pthread
OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 2 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 PGO -O2 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 -O2 -march=athlon64 4 8 12 16 20 SE +/- 0.10, N = 15 SE +/- 0.12, N = 3 SE +/- 0.12, N = 3 SE +/- 0.09, N = 15 SE +/- 0.08, N = 10 SE +/- 0.10, N = 3 SE +/- 0.10, N = 3 14.05 9.59 9.41 9.40 9.01 8.89 8.82 8.68 -O3 -march=native -O3 -march=athlon64 -O3 -march=native -O2 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 -O2 -march=athlon64 1. (CC) gcc options: -pthread
High Performance Conjugate Gradient HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org GFLOP/s, More Is Better High Performance Conjugate Gradient 3.0 -O2 -march=native -O3 -march=native -O3 -march=athlon64-sse3 -O2 -march=athlon64 AMD Ryzen Threadripper 2990WX 32-Core PGO -O3 -march=athlon64 -O3 -march=native -flto 0.2205 0.441 0.6615 0.882 1.1025 SE +/- 0.02, N = 15 SE +/- 0.03, N = 12 SE +/- 0.01, N = 3 SE +/- 0.02, N = 13 SE +/- 0.03, N = 12 SE +/- 0.02, N = 15 SE +/- 0.02, N = 3 0.81 0.83 0.85 0.89 0.90 0.91 0.92 0.98
FFTW FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Stock - Size: 2D FFT Size 4096 -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native PGO -O3 -march=native -flto 1500 3000 4500 6000 7500 SE +/- 28.80, N = 3 SE +/- 63.69, N = 3 SE +/- 1.73, N = 3 SE +/- 34.65, N = 3 SE +/- 51.66, N = 3 SE +/- 14.46, N = 3 SE +/- 8.84, N = 3 4342 4361 4463 6335 6411 6600 6717 7019 -O2 -O3 -O3 -O3 -O3 -flto 1. (CC) gcc options: -pthread -march=native -lm
OpenBenchmarking.org Mflops, More Is Better FFTW 3.3.6 Build: Float + SSE - Size: 2D FFT Size 4096 -O3 -march=native -flto -O3 -march=native AMD Ryzen Threadripper 2990WX 32-Core PGO -O2 -march=native 3K 6K 9K 12K 15K SE +/- 74.46, N = 3 SE +/- 71.85, N = 3 SE +/- 152.85, N = 3 SE +/- 13.87, N = 3 14864 14927 15098 15287 16263 -O3 -flto -O3 -O3 -O3 -O2 1. (CC) gcc options: -pthread -march=native -lm
Timed MAFFT Alignment This test performs an alignment of 100 pyruvate decarboxylase sequences. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed MAFFT Alignment 7.392 Multiple Sequence Alignment AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -O3 -march=native -flto PGO -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 0.6165 1.233 1.8495 2.466 3.0825 SE +/- 0.01, N = 3 SE +/- 0.03, N = 15 SE +/- 0.00, N = 3 SE +/- 0.05, N = 15 SE +/- 0.06, N = 15 SE +/- 0.02, N = 15 SE +/- 0.03, N = 15 2.74 2.69 2.65 2.63 2.60 2.58 2.58 2.55 1. (CC) gcc options: -std=c99 -O3 -lm -lpthread
LuaJIT This test profile is a collection of Lua scripts/benchmarks run against a locally-built copy of LuaJIT upstream. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Composite -O2 -march=native -O3 -march=athlon64 -O2 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 -O3 -march=native -flto PGO 300 600 900 1200 1500 SE +/- 10.67, N = 3 SE +/- 4.29, N = 3 SE +/- 4.50, N = 3 SE +/- 0.72, N = 3 SE +/- 1.46, N = 3 SE +/- 0.43, N = 3 SE +/- 1.60, N = 3 1467 1489 1491 1495 1496 1497 1499 -march=native -O3 -march=athlon64 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Monte Carlo -O2 -march=native -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native PGO -O3 -march=native -flto 110 220 330 440 550 SE +/- 2.87, N = 3 SE +/- 0.13, N = 3 SE +/- 0.46, N = 3 SE +/- 0.04, N = 3 SE +/- 0.17, N = 3 SE +/- 0.60, N = 3 SE +/- 1.65, N = 3 490 498 498 499 499 499 500 -march=native -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -O3 -march=native -flto 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Fast Fourier Transform -O2 -march=native -O2 -march=athlon64 -O3 -march=native PGO -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -flto 60 120 180 240 300 SE +/- 2.77, N = 3 SE +/- 0.26, N = 3 SE +/- 0.06, N = 3 SE +/- 0.27, N = 3 SE +/- 0.62, N = 3 SE +/- 0.81, N = 3 SE +/- 0.09, N = 3 281 286 286 286 287 287 287 -march=native -march=athlon64 -O3 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -flto 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Sparse Matrix Multiply -O2 -march=native -O2 -march=athlon64 -O3 -march=athlon64 PGO -O3 -march=native -flto -O3 -march=athlon64-sse3 -O3 -march=native 300 600 900 1200 1500 SE +/- 7.82, N = 3 SE +/- 14.55, N = 3 SE +/- 2.17, N = 3 SE +/- 0.69, N = 3 SE +/- 1.60, N = 3 SE +/- 1.43, N = 3 SE +/- 1.12, N = 3 1182 1183 1200 1203 1204 1207 1208 -march=native -march=athlon64 -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 -O3 -march=native 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Dense LU Matrix Factorization -O2 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O3 -march=native -flto PGO 800 1600 2400 3200 4000 SE +/- 31.00, N = 3 SE +/- 22.76, N = 3 SE +/- 2.52, N = 3 SE +/- 5.63, N = 3 SE +/- 8.06, N = 3 SE +/- 1.48, N = 3 SE +/- 8.61, N = 3 3550 3590 3611 3619 3623 3624 3636 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 -march=athlon64 -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Jacobi Successive Over-Relaxation -O2 -march=native -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto PGO 400 800 1200 1600 2000 SE +/- 11.29, N = 3 SE +/- 0.07, N = 3 SE +/- 0.46, N = 3 SE +/- 0.46, N = 3 SE +/- 0.27, N = 3 SE +/- 0.35, N = 3 SE +/- 0.37, N = 3 1830 1865 1867 1868 1868 1868 1870 -march=native -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
SciMark This test runs the ANSI C version of SciMark 2.0, which is a benchmark for scientific and numerical computing developed by programmers at the National Institute of Standards and Technology. This test is made up of Fast Foruier Transform, Jacobi Successive Over-relaxation, Monte Carlo, Sparse Matrix Multiply, and dense LU matrix factorization benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Composite -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -O3 -march=native -flto PGO 500 1000 1500 2000 2500 SE +/- 4.62, N = 3 SE +/- 17.63, N = 3 SE +/- 25.27, N = 5 SE +/- 0.49, N = 3 SE +/- 26.08, N = 8 SE +/- 32.19, N = 3 SE +/- 2.91, N = 3 1782 1981 2021 2039 2257 2514 2543 2555 -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Monte Carlo AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O2 -march=athlon64 PGO -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -flto 400 800 1200 1600 2000 SE +/- 3.93, N = 3 SE +/- 3.18, N = 3 SE +/- 0.27, N = 3 SE +/- 0.10, N = 3 SE +/- 0.82, N = 3 SE +/- 0.20, N = 3 SE +/- 0.38, N = 3 255 721 723 728 732 736 737 1904 -O3 -march=native -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -flto 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Fast Fourier Transform AMD Ryzen Threadripper 2990WX 32-Core PGO -O2 -march=native -O3 -march=native -O3 -march=native -flto -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 60 120 180 240 300 SE +/- 0.21, N = 3 SE +/- 1.28, N = 3 SE +/- 0.11, N = 3 SE +/- 0.17, N = 3 SE +/- 0.93, N = 3 SE +/- 0.13, N = 3 SE +/- 0.67, N = 3 260 261 265 270 270 291 294 294 -O3 -march=native -O3 -march=native -O2 -march=native -O3 -march=native -O3 -march=native -flto -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Sparse Matrix Multiply -O3 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native PGO 700 1400 2100 2800 3500 SE +/- 208.10, N = 3 SE +/- 20.52, N = 3 SE +/- 4.22, N = 3 SE +/- 52.49, N = 3 SE +/- 7.39, N = 3 SE +/- 37.52, N = 3 SE +/- 9.00, N = 3 2874 2951 3082 3105 3119 3153 3174 3220 -O3 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=native -O3 -march=native 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Dense LU Matrix Factorization -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native -O3 -march=native -flto AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native PGO 1400 2800 4200 5600 7000 SE +/- 15.82, N = 3 SE +/- 4.78, N = 3 SE +/- 1.14, N = 3 SE +/- 35.70, N = 3 SE +/- 139.92, N = 3 SE +/- 352.63, N = 3 SE +/- 11.42, N = 3 3593 4239 4274 4507 5388 5429 5989 6356 -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native -O3 -march=native -flto -O3 -march=native -O3 -march=native -O3 -march=native 1. (CC) gcc options: -lm
OpenBenchmarking.org Mflops, More Is Better SciMark 2.0 Computational Test: Jacobi Successive Over-Relaxation -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -flto PGO -O3 -march=native 500 1000 1500 2000 2500 SE +/- 5.74, N = 3 SE +/- 7.73, N = 3 SE +/- 0.24, N = 3 SE +/- 0.37, N = 3 SE +/- 0.48, N = 3 SE +/- 0.39, N = 3 SE +/- 0.57, N = 3 1186 1306 1842 1842 2190 2202 2208 2218 -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto -O3 -march=native -O3 -march=native 1. (CC) gcc options: -lm
TSCP This is a performance test of TSCP, Tom Kerrigan's Simple Chess Program, which has a built-in performance benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better TSCP 1.81 AI Chess Performance AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O2 -march=native PGO -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto 200K 400K 600K 800K 1000K SE +/- 5126.15, N = 5 SE +/- 2140.82, N = 5 SE +/- 2184.70, N = 5 SE +/- 1105.70, N = 5 SE +/- 903.00, N = 5 SE +/- 1107.42, N = 5 SE +/- 740.45, N = 5 981778 1094211 1102013 1109114 1115390 1115841 1116747 1135626 -O2 -march=athlon64 -O2 -march=athlon64 -march=athlon64-sse3 -flto 1. (CC) gcc options: -O3 -march=native
MKL-DNN This is a test of the Intel MKL-DNN as the Intel Math Kernel Library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better MKL-DNN 2019-04-16 Harness: IP Batch 1D - Data Type: f32 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core PGO -O3 -march=athlon64 16 32 48 64 80 SE +/- 0.24, N = 3 SE +/- 0.09, N = 3 SE +/- 0.40, N = 3 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.14, N = 3 72.08 71.98 71.26 71.02 71.00 70.94 70.94 -O2 - MIN: 70.86 -O2 -march=athlon64 - MIN: 70.78 MIN: 70.22 -march=athlon64-sse3 - MIN: 70.33 -lrt - MIN: 70.47 MIN: 70.39 -march=athlon64 - MIN: 70.24 1. (CXX) g++ options: -march=native -std=c++11 -mtune=native -fPIC -fopenmp -O3 -pie -lmklml_intel -ldl
AOM AV1 This is a simple test of the AOMedia AV1 encoder run on the CPU with a sample video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 2019-02-11 AV1 Video Encoding AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O3 -march=native -O3 -march=native -flto PGO 0.0495 0.099 0.1485 0.198 0.2475 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 0.08 0.20 0.20 0.21 0.21 0.22 0.22 0.22 -march=native -O2 -march=athlon64 -march=athlon64 -march=athlon64-sse3 -O2 -march=native -march=native -march=native -flto -march=native 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread
SVT-AV1 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 2019-03-07 1080p 8-bit YUV To AV1 Video Encode -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 PGO -O3 -march=native -O3 -march=native -flto 5 10 15 20 25 SE +/- 0.17, N = 3 SE +/- 0.08, N = 3 SE +/- 0.09, N = 3 SE +/- 0.11, N = 3 SE +/- 0.10, N = 3 SE +/- 0.05, N = 3 SE +/- 0.08, N = 3 18.51 18.63 18.66 18.77 18.84 19.57 20.27 20.41 -O2 -march=athlon64 -O2 -march=native -march=athlon64-sse3 -march=native -march=athlon64 -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -pie -lpthread -lm
SVT-HEVC This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 PGO 40 80 120 160 200 SE +/- 3.04, N = 12 SE +/- 1.96, N = 3 SE +/- 2.04, N = 3 SE +/- 0.65, N = 3 SE +/- 2.41, N = 3 SE +/- 4.80, N = 15 SE +/- 6.47, N = 15 23.01 163.00 165.00 165.00 166.00 168.00 172.00 185.00 -O3 -O3 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -march=athlon64 -O3 1. (CC) gcc options: -march=native -fPIE -fPIC -O2 -flto -fvisibility=hidden -pie -rdynamic -lpthread -lrt
SVT-VP9 This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=athlon64 PGO -O3 -march=native -flto 20 40 60 80 100 SE +/- 0.78, N = 3 SE +/- 0.82, N = 3 SE +/- 0.98, N = 15 SE +/- 1.13, N = 15 SE +/- 1.26, N = 15 SE +/- 0.96, N = 15 SE +/- 1.19, N = 15 5.07 97.81 100.82 101.13 102.91 103.45 103.97 104.42 -O3 -march=athlon64-sse3 -march=native -march=athlon64 -O3 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=native 1. (CC) gcc options: -fPIE -fPIC -O2 -flto -fvisibility=hidden -mavx -pie -rdynamic -lpthread -lrt -lm
VP9 libvpx Encoding This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format using a sample 1080p video. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O3 -march=athlon64 PGO -O3 -march=native 6 12 18 24 30 SE +/- 0.26, N = 9 SE +/- 0.40, N = 3 SE +/- 0.22, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.13, N = 3 25.44 25.58 26.02 26.12 26.36 26.37 -O2 -march=athlon64 -march=athlon64-sse3 -O2 -march=native -march=athlon64 -march=native -march=native 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
x264 This is a simple test of the x264 encoder run on the CPU (OpenCL support disabled) with a sample video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better x264 2018-09-25 H.264 Video Encoding -O2 -march=native PGO -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native 30 60 90 120 150 SE +/- 1.95, N = 3 SE +/- 0.95, N = 3 SE +/- 1.63, N = 7 SE +/- 0.81, N = 3 SE +/- 1.95, N = 5 SE +/- 1.46, N = 9 143 145 146 146 146 147 -O2 -march=native -march=native -O2 -march=athlon64 -march=athlon64 -march=athlon64-sse3 -march=native 1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fPIC -fomit-frame-pointer -fno-tree-vectorize
x265 This is a simple test of the x265 encoder run on the CPU with a sample 1080p video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better x265 3.0 H.265 1080p Video Encoding AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O2 -march=athlon64 -O3 -march=native -flto -O3 -march=native PGO -O3 -march=athlon64-sse3 -O3 -march=athlon64 8 16 24 32 40 SE +/- 0.09, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.09, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 9.95 33.44 33.53 33.68 33.76 33.79 33.88 33.89 -march=native -O2 -march=native -O2 -march=athlon64 -march=native -flto -march=native -march=native -march=athlon64-sse3 -march=athlon64 1. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma
GraphicsMagick This is a test of GraphicsMagick with its OpenMP implementation that performs various imaging tests to stress the system's CPU. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Swirl AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=native -flto PGO 50 100 150 200 250 SE +/- 0.58, N = 3 SE +/- 0.33, N = 3 SE +/- 1.00, N = 3 26 237 237 244 245 247 250 250 -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Rotate AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O3 -march=native -flto -O2 -march=native -O3 -march=native PGO 50 100 150 200 250 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 175 238 240 243 248 249 249 251 -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O3 -march=native -flto -O2 -march=native -O3 -march=native -O3 -march=native 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Sharpen AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O2 -march=native -O3 -march=native PGO -O3 -march=native -flto 50 100 150 200 250 SE +/- 0.67, N = 3 SE +/- 0.67, N = 3 SE +/- 0.33, N = 3 SE +/- 0.88, N = 3 SE +/- 0.88, N = 3 4 190 191 192 217 219 220 221 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=native -O3 -march=native -flto 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Enhanced AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=native -flto PGO 50 100 150 200 250 SE +/- 0.58, N = 3 SE +/- 0.58, N = 3 SE +/- 0.88, N = 3 8 194 195 198 231 232 233 234 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Resizing AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -flto -O3 -march=native PGO 50 100 150 200 250 SE +/- 1.15, N = 3 SE +/- 0.88, N = 3 SE +/- 1.00, N = 3 SE +/- 0.67, N = 3 31 231 233 235 238 240 243 243 -O3 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -flto -O3 -march=native -O3 -march=native 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Noise-Gaussian AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -flto PGO -O3 -march=native 40 80 120 160 200 SE +/- 0.88, N = 3 SE +/- 0.67, N = 3 SE +/- 1.53, N = 3 SE +/- 0.67, N = 3 SE +/- 0.67, N = 3 40 198 199 200 202 203 203 204 -O3 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -flto -O3 -march=native -O3 -march=native 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: HWB Color Space AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=native -flto PGO 60 120 180 240 300 SE +/- 1.33, N = 3 SE +/- 0.33, N = 3 59 261 263 271 272 272 274 274 -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=native -flto -O3 -march=native 1. (CC) gcc options: -fopenmp -pthread -ljbig -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lz -lm -ldl -lpthread
Himeno Benchmark The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver -O3 -march=native -flto -O3 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native PGO AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 300 600 900 1200 1500 SE +/- 1.86, N = 3 SE +/- 1.73, N = 3 SE +/- 3.98, N = 3 SE +/- 4.00, N = 3 SE +/- 3.88, N = 3 SE +/- 3.47, N = 3 SE +/- 3.65, N = 3 1304 1313 1316 1316 1319 1321 1322 1328 -march=native -flto -march=native -march=athlon64 -march=athlon64-sse3 -O2 -march=native -march=native -march=native -O2 -march=athlon64 1. (CC) gcc options: -O3 -mavx2
Stockfish This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time -O2 -march=native -O2 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64 -O3 -march=athlon64-sse3 PGO -O3 -march=native 15M 30M 45M 60M 75M SE +/- 526520.80, N = 3 SE +/- 226981.30, N = 3 SE +/- 385955.10, N = 3 SE +/- 779150.59, N = 3 SE +/- 443906.77, N = 3 SE +/- 458502.78, N = 3 SE +/- 954655.13, N = 3 66697487 66890687 67450689 67513602 67571150 67841877 68200164 -O2 -march=native -O2 -march=athlon64 -march=native -march=athlon64 -march=athlon64-sse3 -march=native -march=native 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto
Timed ImageMagick Compilation This test times how long it takes to build ImageMagick. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed ImageMagick Compilation 6.9.0 Time To Compile -O3 -march=native -flto -O3 -march=athlon64 -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native PGO -O2 -march=native -O2 -march=athlon64 20 40 60 80 100 SE +/- 1.39, N = 3 SE +/- 0.23, N = 5 SE +/- 0.03, N = 3 SE +/- 0.25, N = 3 SE +/- 0.07, N = 3 SE +/- 0.24, N = 3 SE +/- 0.21, N = 3 88.89 19.77 19.73 19.69 19.15 19.09 17.58 17.41
Timed PHP Compilation This test times how long it takes to build PHP 5 with the Zend engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed PHP Compilation 7.1.9 Time To Compile AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -O3 -march=athlon64 PGO -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 20 40 60 80 100 SE +/- 0.08, N = 3 SE +/- 0.32, N = 3 SE +/- 0.10, N = 3 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.20, N = 3 85.40 63.25 63.19 63.01 62.93 44.44 43.86 -O3 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 -O2 -march=native -O2 -march=athlon64 1. (CC) gcc options: -pedantic -ldl -lz -lm
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O3 -march=native PGO -O3 -march=native -flto 900 1800 2700 3600 4500 SE +/- 0.13, N = 3 SE +/- 0.12, N = 3 SE +/- 0.05, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 4024.62 45.43 34.22 29.53 29.39 18.00 17.96 17.85 -march=native 1. (CC) gcc options: -lm -lpthread -O3
Smallpt Smallpt is a C++ global illumination renderer written in less than 100 lines of code. Global illumination is done via unbiased Monte Carlo path tracing and there is multi-threading support via the OpenMP library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Smallpt 1.0 Global Illumination Renderer; 128 Samples AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=native -flto PGO 120 240 360 480 600 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 564.07 4.57 4.54 4.51 3.89 3.87 3.85 3.83 -march=native 1. (CXX) g++ options: -fopenmp -O3
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O3 -march=native -flto -O3 -march=native PGO 11 22 33 44 55 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 47.28 45.20 44.98 43.36 42.34 39.92 39.11 39.10 -O2 -march=athlon64 -march=athlon64 -march=athlon64-sse3 -march=native -O2 -march=native -march=native -flto -march=native -march=native 1. (CC) gcc options: -lm -O3
Bullet Physics Engine This is a benchmark of the Bullet Physics Engine. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Raytests -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native -O3 -march=native PGO -O3 -march=native -flto 0.5603 1.1206 1.6809 2.2412 2.8015 SE +/- 0.02, 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 2.49 2.46 2.46 2.38 2.35 2.34 2.30 -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -O2 -march=native -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 3000 Fall -O3 -march=native -flto -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native PGO 0.8933 1.7866 2.6799 3.5732 4.4665 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 3.97 3.96 3.95 3.92 3.91 3.84 3.83 -march=native -flto -O2 -march=athlon64 -O2 -march=native -march=athlon64 -march=athlon64-sse3 -march=native -march=native 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Stack -O3 -march=native -flto -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O3 -march=native PGO 1.0845 2.169 3.2535 4.338 5.4225 SE +/- 0.01, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 4.82 4.68 4.65 4.62 4.57 4.42 4.41 -march=native -flto -O2 -march=athlon64 -march=athlon64 -march=athlon64-sse3 -O2 -march=native -march=native -march=native 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 1000 Convex -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native PGO -O3 -march=native -O3 -march=native -flto 1.0125 2.025 3.0375 4.05 5.0625 SE +/- 0.04, N = 3 SE +/- 0.01, 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 4.50 4.45 4.45 4.06 3.97 3.97 3.90 -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -O2 -march=native -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: 136 Ragdolls -O3 -march=native -flto -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 PGO -O3 -march=native 0.549 1.098 1.647 2.196 2.745 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 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 2.44 2.42 2.39 2.36 2.35 2.32 2.30 -march=native -flto -O2 -march=athlon64 -O2 -march=native -march=athlon64 -march=athlon64-sse3 -march=native -march=native 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Prim Trimesh -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O3 -march=native -O2 -march=native PGO -O3 -march=native -flto 0.1958 0.3916 0.5874 0.7832 0.979 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 0.87 0.86 0.86 0.84 0.84 0.83 0.82 -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -march=native -O2 -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
OpenBenchmarking.org Seconds, Fewer Is Better Bullet Physics Engine 2.81 Test: Convex Trimesh -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=native PGO -O3 -march=native -O3 -march=native -flto 0.2408 0.4816 0.7224 0.9632 1.204 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 1.07 1.06 1.06 1.02 1.00 1.00 0.97 -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -O2 -march=native -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU
XZ Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using XZ compression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O2 -march=athlon64 -O3 -march=athlon64 PGO -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto 150 300 450 600 750 SE +/- 0.10, N = 3 SE +/- 0.21, N = 15 SE +/- 0.27, N = 3 SE +/- 0.07, N = 3 SE +/- 0.22, N = 12 SE +/- 0.26, N = 15 SE +/- 0.23, N = 15 708.15 26.78 26.75 26.50 26.09 25.93 25.70 25.62 -O3 -march=native 1. (CC) gcc options: -pthread -fvisibility=hidden
Zstd Compression This test measures the time needed to compress a sample file (an Ubuntu file-system image) using Zstd compression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O3 -march=native -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64 PGO 60 120 180 240 300 SE +/- 0.61, N = 15 SE +/- 0.65, N = 15 SE +/- 0.66, N = 15 SE +/- 0.60, N = 15 SE +/- 0.66, N = 15 SE +/- 0.69, N = 12 SE +/- 0.45, N = 15 265.79 19.31 19.09 18.38 18.02 17.94 17.74 17.36 -O3 -march=native 1. (CC) gcc options: -pthread -lz -llzma
FLAC Audio Encoding This test times how long it takes to encode a sample WAV file to FLAC format five times. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better FLAC Audio Encoding 1.3.2 WAV To FLAC -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O3 -march=native -O3 -march=native -flto PGO 4 8 12 16 20 SE +/- 0.01, N = 5 SE +/- 0.04, N = 5 SE +/- 0.01, N = 5 SE +/- 0.05, N = 5 SE +/- 0.01, N = 5 SE +/- 0.06, N = 5 SE +/- 0.00, N = 5 15.58 15.48 15.43 9.81 9.78 9.53 9.48 9.44 -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O2 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=native 1. (CXX) g++ options: -fvisibility=hidden -lm
LAME MP3 Encoding LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better LAME MP3 Encoding 3.100 WAV To MP3 -O2 -march=athlon64 AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native PGO -O3 -march=native -flto 3 6 9 12 15 SE +/- 0.04, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 11.41 11.09 10.80 9.43 9.36 8.00 7.98 7.98 -O2 -march=athlon64 -O3 -march=native -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -O3 -march=native -flto 1. (CC) gcc options: -lm
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native PGO -O2 -march=native -O2 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64 -O3 -march=athlon64-sse3 1300 2600 3900 5200 6500 SE +/- 3.74, N = 3 SE +/- 4.62, N = 3 SE +/- 1.56, N = 3 SE +/- 0.92, N = 3 SE +/- 2.44, N = 3 SE +/- 1.40, N = 3 SE +/- 1.25, N = 3 5791 5825 5830 5831 5832 5833 5837 5838 -O3 -march=native -O3 -march=native -lssl -O3 -march=native -lssl -O2 -march=native -lssl -O2 -march=athlon64 -lssl -O3 -march=native -flto -lssl -O3 -march=athlon64 -lssl -O3 -march=athlon64-sse3 -lssl 1. (CC) gcc options: -pthread -m64 -lcrypto -ldl
PostgreSQL pgbench This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64-sse3 PGO -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto 100K 200K 300K 400K 500K SE +/- 2874.20, N = 3 SE +/- 3403.88, N = 3 SE +/- 6062.49, N = 4 SE +/- 7464.16, N = 3 SE +/- 6987.31, N = 3 SE +/- 1148.20, N = 3 SE +/- 1547.77, N = 3 259058 453605 458551 459596 460414 463840 466723 473620 -O3 -march=native -O2 -march=athlon64 -lpq -O2 -march=native -lpq -O3 -march=athlon64-sse3 -lpq -O3 -march=native -lpq -O3 -march=athlon64 -lpq -O3 -march=native -lpq -O3 -march=native -flto -lpq 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write -O2 -march=athlon64 PGO AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O3 -march=native -flto -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native 4K 8K 12K 16K 20K SE +/- 65.89, N = 4 SE +/- 93.40, N = 4 SE +/- 279.82, N = 15 SE +/- 164.04, N = 7 SE +/- 247.31, N = 3 SE +/- 51.55, N = 3 SE +/- 98.94, N = 3 5098 6281 6472 13558 14989 15616 16252 16339 -O2 -march=athlon64 -lpq -O3 -march=native -lpq -O3 -march=native -O2 -march=native -lpq -O3 -march=native -flto -lpq -O3 -march=athlon64 -lpq -O3 -march=athlon64-sse3 -lpq -O3 -march=native -lpq 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm
CppPerformanceBenchmarks CppPerformanceBenchmarks is a set of C++ compiler performance benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Atol -O2 -march=native PGO AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -flto -O2 -march=athlon64 -O3 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 16 32 48 64 80 SE +/- 0.13, N = 3 SE +/- 0.17, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.38, N = 3 SE +/- 0.39, N = 3 69.80 69.31 69.25 69.25 69.08 69.03 68.42 68.37 -O2 -march=native -march=native -march=native -march=native -flto -O2 -march=athlon64 -march=native -march=athlon64-sse3 -march=athlon64 1. (CXX) g++ options: -std=c++11 -O3
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Ctype AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native PGO -O3 -march=native -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O3 -march=native -flto 9 18 27 36 45 SE +/- 0.08, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 37.67 34.50 34.02 33.88 33.37 33.31 33.28 32.32 -march=native -O2 -march=native -march=native -march=native -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -march=native -flto 1. (CXX) g++ options: -O3 -std=c++11
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Math Library -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O3 -march=athlon64 -O2 -march=native AMD Ryzen Threadripper 2990WX 32-Core PGO -O3 -march=native -flto -O3 -march=native 90 180 270 360 450 SE +/- 0.83, N = 3 SE +/- 0.87, N = 3 SE +/- 0.79, N = 3 SE +/- 1.79, N = 3 SE +/- 0.93, N = 3 SE +/- 0.73, N = 3 SE +/- 0.11, N = 3 399 399 398 356 355 353 352 351 -march=athlon64-sse3 -O2 -march=athlon64 -march=athlon64 -O2 -march=native -march=native -march=native -march=native -flto -march=native 1. (CXX) g++ options: -O3 -std=c++11
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Random Numbers AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native PGO -O3 -march=native -O3 -march=native -flto 200 400 600 800 1000 SE +/- 0.07, N = 3 SE +/- 0.71, N = 3 SE +/- 0.04, N = 3 SE +/- 4.70, N = 3 SE +/- 0.03, N = 3 SE +/- 0.06, N = 3 SE +/- 0.05, N = 3 1154 1072 1070 1055 1041 1027 1023 1011 -march=native -O2 -march=athlon64 -march=athlon64 -march=athlon64-sse3 -O2 -march=native -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -std=c++11
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Vector AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 PGO -O3 -march=native -O3 -march=native -flto 20 40 60 80 100 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 78.14 75.99 75.79 75.56 75.53 75.27 74.93 74.92 -march=native -O2 -march=native -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -march=native -march=native -march=native -flto 1. (CXX) g++ options: -O3 -std=c++11
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Function Objects AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=athlon64 -O3 -march=athlon64 -O3 -march=athlon64-sse3 -O2 -march=native -O3 -march=native -flto PGO -O3 -march=native 5 10 15 20 25 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 22.07 16.20 16.17 16.14 15.61 15.50 15.46 15.40 -march=native -O2 -march=athlon64 -march=athlon64 -march=athlon64-sse3 -O2 -march=native -march=native -flto -march=native -march=native 1. (CXX) g++ options: -O3 -std=c++11
OpenBenchmarking.org Seconds, Fewer Is Better CppPerformanceBenchmarks 9 Test: Stepanov Abstraction AMD Ryzen Threadripper 2990WX 32-Core -O2 -march=native PGO -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O3 -march=native -flto -O3 -march=native 7 14 21 28 35 SE +/- 0.13, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 29.08 28.69 28.36 28.33 28.28 28.26 28.25 28.22 -march=native -O2 -march=native -march=native -O2 -march=athlon64 -march=athlon64-sse3 -march=athlon64 -march=native -flto -march=native 1. (CXX) g++ options: -O3 -std=c++11
Redis Redis is an open-source data structure server. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPOP -O2 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto PGO -O2 -march=athlon64 600K 1200K 1800K 2400K 3000K SE +/- 24290.24, N = 3 SE +/- 3626.56, N = 3 SE +/- 25886.44, N = 8 SE +/- 34288.12, N = 3 SE +/- 24550.99, N = 9 SE +/- 25803.36, N = 3 SE +/- 41423.26, N = 15 2340595 2506276 2544762 2573823 2616703 2648345 2654025 -O2 -O3 -march=native -flto 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SADD -O2 -march=athlon64 -O2 -march=native -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=athlon64 PGO -O3 -march=native -flto 400K 800K 1200K 1600K 2000K SE +/- 18375.04, N = 3 SE +/- 26134.11, N = 12 SE +/- 24795.31, N = 5 SE +/- 13234.25, N = 3 SE +/- 23560.29, N = 3 SE +/- 10024.22, N = 3 SE +/- 27875.47, N = 3 1975322 2000709 2032921 2046557 2051129 2083430 2084089 -O2 -O3 -march=native -flto 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: LPUSH -O2 -march=native -O3 -march=athlon64 -O3 -march=athlon64-sse3 PGO -O3 -march=native -O2 -march=athlon64 -O3 -march=native -flto 300K 600K 900K 1200K 1500K SE +/- 22581.44, N = 3 SE +/- 19988.57, N = 3 SE +/- 9157.67, N = 3 SE +/- 22070.89, N = 3 SE +/- 12857.88, N = 3 SE +/- 20493.00, N = 5 SE +/- 6810.29, N = 3 1463388 1520276 1522952 1532038 1540257 1548556 1549647 -O2 -O3 -march=native -flto 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: GET -O2 -march=native -O2 -march=athlon64 -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -flto PGO -O3 -march=athlon64 500K 1000K 1500K 2000K 2500K SE +/- 25441.63, N = 3 SE +/- 39327.40, N = 3 SE +/- 28994.69, N = 3 SE +/- 11016.75, N = 3 SE +/- 36774.66, N = 3 SE +/- 26635.52, N = 12 SE +/- 15021.00, N = 3 2275019 2384143 2461725 2502182 2509433 2533027 2540398 -O2 -O3 -march=native -flto 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
OpenBenchmarking.org Requests Per Second, More Is Better Redis 4.0.8 Test: SET -O2 -march=athlon64 -O2 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 PGO -O3 -march=native -O3 -march=athlon64 400K 800K 1200K 1600K 2000K SE +/- 22619.48, N = 3 SE +/- 28790.51, N = 15 SE +/- 7219.96, N = 3 SE +/- 12476.39, N = 3 SE +/- 14705.69, N = 3 SE +/- 6043.61, N = 3 SE +/- 16989.46, N = 3 1730702 1744979 1755472 1755590 1798800 1807270 1840806 -O2 -O3 -march=native -flto 1. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread
ctx_clock Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time AMD Ryzen Threadripper 2990WX 32-Core PGO -O3 -march=native -flto -O3 -march=native -O2 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 30 60 90 120 150 150 150 150 150 150 150 150 150 -O3 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=native -O2 -march=native -O3 -march=athlon64-sse3 -O3 -march=athlon64 -O2 -march=athlon64 1. (CC) gcc options:
Memcached mcperf This is a test of twmperf/mcperf with memcached. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Add -O2 -march=native -O3 -march=athlon64-sse3 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native PGO -O3 -march=native -flto -O3 -march=athlon64 -O2 -march=athlon64 12K 24K 36K 48K 60K SE +/- 125.83, N = 3 SE +/- 121.25, N = 3 SE +/- 1053.86, N = 15 SE +/- 2440.20, N = 12 SE +/- 2185.23, N = 15 SE +/- 2252.95, N = 15 SE +/- 2866.90, N = 12 34138 34822 35174 44903 47774 50055 53106 54600 -O2 -march=native -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64 -O2 -march=athlon64 1. (CC) gcc options: -lm -rdynamic
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Get -O2 -march=athlon64 -O2 -march=native AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=native PGO -O3 -march=native -flto -O3 -march=athlon64-sse3 15K 30K 45K 60K 75K SE +/- 339.02, N = 3 SE +/- 125.71, N = 3 SE +/- 29.60, N = 3 SE +/- 812.95, N = 3 SE +/- 734.34, N = 3 SE +/- 66.15, N = 3 SE +/- 286.25, N = 3 55647 55791 56324 56837 57652 68426 68644 69004 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 1. (CC) gcc options: -lm -rdynamic
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Set -O2 -march=native -O3 -march=athlon64 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -O3 -march=native -flto PGO -O3 -march=athlon64-sse3 -O2 -march=athlon64 13K 26K 39K 52K 65K SE +/- 67.69, N = 3 SE +/- 44.53, N = 3 SE +/- 877.57, N = 15 SE +/- 195.82, N = 3 SE +/- 1474.59, N = 12 SE +/- 2384.39, N = 15 SE +/- 2193.95, N = 15 34446 34880 35239 38646 43570 46396 48548 59924 -O2 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=native -O3 -march=athlon64-sse3 -O2 -march=athlon64 1. (CC) gcc options: -lm -rdynamic
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Append -O2 -march=native AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 PGO -O3 -march=athlon64 -O2 -march=athlon64 13K 26K 39K 52K 65K SE +/- 168.27, N = 3 SE +/- 256.24, N = 3 SE +/- 218.35, N = 3 SE +/- 220.19, N = 3 SE +/- 123.87, N = 3 SE +/- 3045.08, N = 15 SE +/- 2658.53, N = 15 35058 35920 42455 45339 45716 46106 56172 61219 -O2 -march=native -O3 -march=native -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=athlon64 -O2 -march=athlon64 1. (CC) gcc options: -lm -rdynamic
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Delete -O2 -march=athlon64 -O2 -march=native AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto PGO -O3 -march=athlon64-sse3 15K 30K 45K 60K 75K SE +/- 274.96, N = 3 SE +/- 304.30, N = 3 SE +/- 204.60, N = 3 SE +/- 772.91, N = 4 SE +/- 220.33, N = 3 SE +/- 370.63, N = 3 SE +/- 914.89, N = 4 55747 56141 56797 56822 58969 68696 68980 69509 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto -O3 -march=native -O3 -march=athlon64-sse3 1. (CC) gcc options: -lm -rdynamic
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Prepend -O2 -march=athlon64 -O2 -march=native AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 PGO 10K 20K 30K 40K 50K SE +/- 109.24, N = 3 SE +/- 186.47, N = 3 SE +/- 58.44, N = 3 SE +/- 2300.04, N = 15 SE +/- 179.60, N = 3 SE +/- 116.90, N = 3 SE +/- 863.93, N = 12 35467 35552 35579 35968 43824 45587 45691 47085 -O2 -march=athlon64 -O2 -march=native -O3 -march=native -O3 -march=athlon64 -O3 -march=native -O3 -march=native -flto -O3 -march=athlon64-sse3 -O3 -march=native 1. (CC) gcc options: -lm -rdynamic
OpenBenchmarking.org Operations Per Second, More Is Better Memcached mcperf 1.5.10 Method: Replace -O2 -march=native -O2 -march=athlon64 AMD Ryzen Threadripper 2990WX 32-Core -O3 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64-sse3 PGO -O3 -march=native 12K 24K 36K 48K 60K SE +/- 167.62, N = 3 SE +/- 132.68, N = 3 SE +/- 365.15, N = 3 SE +/- 174.58, N = 3 SE +/- 85.02, N = 3 SE +/- 168.47, N = 3 SE +/- 1741.69, N = 12 35486 35829 35978 36231 45591 45707 45956 53865 -O2 -march=native -O2 -march=athlon64 -O3 -march=native -O3 -march=athlon64 -O3 -march=native -flto -O3 -march=athlon64-sse3 -O3 -march=native -O3 -march=native 1. (CC) gcc options: -lm -rdynamic
NGINX Benchmark This is a test of ab, which is the Apache Benchmark program running against nginx. This test profile measures how many requests per second a given system can sustain when carrying out 2,000,000 requests with 500 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving -O3 -march=native -flto -O2 -march=native -O3 -march=native -O3 -march=athlon64-sse3 -O2 -march=athlon64 -O3 -march=athlon64 6K 12K 18K 24K 30K SE +/- 43.76, N = 3 SE +/- 166.84, N = 3 SE +/- 401.04, N = 4 SE +/- 428.25, N = 3 SE +/- 219.17, N = 3 SE +/- 169.85, N = 3 27352 27834 29274 29281 29704 29726 -flto -O2 -march=athlon64-sse3 -O2 -march=athlon64 -march=athlon64 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
-O2 -march=athlon64 Environment Notes: CXXFLAGS=-O2-march=athlon64 CFLAGS=-O2-march=athlon64Compiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 11 May 2019 17:06 by user phoronix.
-O3 -march=athlon64 Environment Notes: CXXFLAGS=-O3-march=athlon64 CFLAGS=-O3-march=athlon64Compiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 9 May 2019 13:16 by user phoronix.
-O3 -march=athlon64-sse3 Environment Notes: CXXFLAGS=-O3-march=athlon64-sse3 CFLAGS=-O3-march=athlon64-sse3Compiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 9 May 2019 08:33 by user phoronix.
-O2 -march=native Environment Notes: CXXFLAGS=-O2-march=native CFLAGS=-O2-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 10 May 2019 06:28 by user phoronix.
-O3 -march=native Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 8 May 2019 19:49 by user phoronix.
-O3 -march=native -flto Environment Notes: CXXFLAGS=-O3-march=native-flto CFLAGS=-O3-march=native-fltoCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 9 May 2019 21:12 by user phoronix.
PGO Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 12 May 2019 09:23 by user phoronix.
AMD Ryzen Threadripper 2990WX 32-Core Processor: AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH EXTREME (1701 BIOS), Chipset: AMD 17h, Memory: 32768MB, Disk: Samsung SSD 970 EVO 500GB, Graphics: AMD Radeon RX 64 8GB (1590/800MHz), Audio: Realtek ALC1220, Monitor: ASUS VP28U, Network: Intel I211 + Qualcomm Atheros QCA6174 802.11ac + Wilocity Wil6200 802.11ad
OS: Ubuntu 18.04, Kernel: 4.18.0-18-generic (x86_64), Desktop: GNOME Shell 3.28.3, Display Server: X Server 1.20.1, Display Driver: amdgpu 18.1.0, OpenGL: 4.5 Mesa 18.2.8 (LLVM 7.0.0), Compiler: GCC 9.1.0, File-System: ext4, Screen Resolution: 3840x2160
Environment Notes: CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Notes: --disable-multilib --enable-checing=releaseProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15rc1 + Python 3.6.7Security Notes: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
Testing initiated at 12 May 2019 14:52 by user phoronix.