GCC vs. Clang Linux 5.5 built kernels by Michael Larabel for a future article.
Clang 9.0 Built Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0601 BIOS), Chipset: AMD Starship/Matisse, Memory: 64512MB, Disk: 1000GB Force MP600, Graphics: AMD Radeon VII 16GB (1802/1001MHz), Audio: AMD Vega 20 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 19.10, Kernel: 5.5.0-rc3-clang9phx (x86_64) 20191224, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: amdgpu 19.0.1, OpenGL: 4.5 Mesa 19.2.1 (LLVM 9.0.0), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 3840x2160
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Graphics Notes: GLAMORPython Notes: Python 2.7.17rc1 + Python 3.7.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected
GCC 9.2 Built OS: Ubuntu 19.10, Kernel: 5.5.0-rc3-gcc9phx (x86_64) 20191224, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: amdgpu 19.0.1, OpenGL: 4.5 Mesa 19.2.1 (LLVM 9.0.0), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 3840x2160
SQLite This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.30.1 Threads / Copies: 1 GCC 9.2 Built Clang 9.0 Built 4 8 12 16 20 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 14.33 14.21 1. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.30.1 Threads / Copies: 8 GCC 9.2 Built Clang 9.0 Built 9 18 27 36 45 SE +/- 0.66, N = 3 SE +/- 0.42, N = 3 40.39 40.89 1. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size GCC 9.2 Built Clang 9.0 Built 130 260 390 520 650 SE +/- 18.56, N = 15 SE +/- 4.40, N = 3 549.8 579.2 1. (CC) gcc options: -static
OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size, No Sync/FSync GCC 9.2 Built Clang 9.0 Built 600 1200 1800 2400 3000 SE +/- 13.36, N = 3 SE +/- 18.73, N = 3 2967.5 2926.5 1. (CC) gcc options: -static
BlogBench BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Read GCC 9.2 Built Clang 9.0 Built 300K 600K 900K 1200K 1500K SE +/- 15836.11, N = 9 SE +/- 23808.12, N = 3 1477144 1541614 1. (CC) gcc options: -O2 -pthread
OpenBenchmarking.org Final Score, More Is Better BlogBench 1.1 Test: Write GCC 9.2 Built Clang 9.0 Built 12K 24K 36K 48K 60K SE +/- 3053.28, N = 3 SE +/- 2302.33, N = 3 53281 54915 1. (CC) gcc options: -O2 -pthread
OpenBenchmarking.org MB/s, More Is Better Dbench 4.0 Client Count: 6 Clang 9.0 Built 80 160 240 320 400 SE +/- 21.82, N = 8 351.20 1. (CC) gcc options: -lpopt -O2
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile GCC 9.2 Built Clang 9.0 Built 500 1000 1500 2000 2500 SE +/- 9.36, N = 3 SE +/- 27.97, N = 6 2364.34 2361.80
SuperTuxKart This is a test of SuperTuxKart, an open-source racing game. The engine, Antarctica, is a highly modified version of Irrlicht and uses OpenGL 3.1+. Bullet is used for physics. Sound is provided by OpenAL. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SuperTuxKart 0.9.3 Resolution: 1920 x 1080 - Mode: Fullscreen - Graphics Effects: Ultimate - Karts: 8 - Scene: Candela City GCC 9.2 Built Clang 9.0 Built 50 100 150 200 250 SE +/- 1.47, N = 3 SE +/- 3.00, N = 4 222.14 220.15
OpenBenchmarking.org Frames Per Second, More Is Better SuperTuxKart 0.9.3 Resolution: 1920 x 1080 - Mode: Fullscreen - Graphics Effects: Ultimate - Karts: 8 - Scene: Cocoa Temple GCC 9.2 Built Clang 9.0 Built 40 80 120 160 200 SE +/- 1.91, N = 3 SE +/- 3.26, N = 3 196.29 192.18
OpenBenchmarking.org FPS, More Is Better GeeXLab 0.28.0 Resolution: 3840 x 2160 - Test: GL3 Vertex Pool GCC 9.2 Built Clang 9.0 Built 1400 2800 4200 5600 7000 SE +/- 26.21, N = 3 SE +/- 11.23, N = 3 6634.4 6623.7
OpenBenchmarking.org FPS, More Is Better GeeXLab 0.28.0 Resolution: 3840 x 2160 - Test: GL2 Cell Shading GCC 9.2 Built Clang 9.0 Built 900 1800 2700 3600 4500 SE +/- 4.96, N = 3 SE +/- 13.62, N = 3 3972.9 3982.0
OpenBenchmarking.org FPS, More Is Better GeeXLab 0.28.0 Resolution: 3840 x 2160 - Test: GL2 Tunnel Beauty GCC 9.2 Built Clang 9.0 Built 30 60 90 120 150 SE +/- 0.00, N = 3 SE +/- 0.12, N = 3 112.8 112.1
OpenBenchmarking.org FPS, More Is Better GeeXLab 0.28.0 Resolution: 3840 x 2160 - Test: GL3 Mesh Exploder GCC 9.2 Built Clang 9.0 Built 900 1800 2700 3600 4500 SE +/- 17.25, N = 3 SE +/- 8.78, N = 3 3992.5 4039.1
OpenBenchmarking.org FPS, More Is Better GeeXLab 0.28.0 Resolution: 3840 x 2160 - Test: GL2 Hot Tunnel DNA GCC 9.2 Built Clang 9.0 Built 30 60 90 120 150 SE +/- 0.09, N = 3 SE +/- 0.21, N = 3 136.3 135.7
OpenBenchmarking.org FPS, More Is Better GeeXLab 0.28.0 Resolution: 3840 x 2160 - Test: GL2 Noise Animation Electric GCC 9.2 Built Clang 9.0 Built 600 1200 1800 2400 3000 SE +/- 11.21, N = 3 SE +/- 11.71, N = 3 2694.1 2686.9
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.4.1 Test: Many Spheres - Resolution: 3840 x 2160 GCC 9.2 Built Clang 9.0 Built 900 1800 2700 3600 4500 SE +/- 32.69, N = 3 SE +/- 41.24, N = 3 4403.15 4412.36
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Volume - Resolution: 3840 x 2160 GCC 9.2 Built Clang 9.0 Built 30 60 90 120 150 SE +/- 5.41, N = 12 SE +/- 0.87, N = 3 123.08 117.59
OpenBenchmarking.org MiVoxels / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Volume - Resolution: 3840 x 2160 GCC 9.2 Built Clang 9.0 Built 400 800 1200 1600 2000 SE +/- 86.53, N = 12 SE +/- 13.89, N = 3 1969.29 1881.38
OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Contour - Resolution: 3840 x 2160 GCC 9.2 Built Clang 9.0 Built 40 80 120 160 200 SE +/- 3.11, N = 3 SE +/- 1.32, N = 3 180.87 168.14
OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Contour - Resolution: 3840 x 2160 GCC 9.2 Built Clang 9.0 Built 400 800 1200 1600 2000 SE +/- 32.42, N = 3 SE +/- 13.76, N = 3 1884.83 1752.24
GpuTest GpuTest is a cross-platform OpenGL benchmark developed at Geeks3D.com that offers tech demos such as FurMark, TessMark, and other workloads to stress various areas of GPUs and drivers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: GiMark - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 2K 4K 6K 8K 10K SE +/- 8.50, N = 3 SE +/- 16.17, N = 3 7894 7887
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: Plot3D - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 20K 40K 60K 80K 100K SE +/- 1.00, N = 3 95816 95793
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: Furmark - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 1300 2600 3900 5200 6500 SE +/- 53.69, N = 3 SE +/- 68.61, N = 3 6000 5944
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: TessMark - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 3K 6K 9K 12K 15K SE +/- 15.71, N = 3 SE +/- 41.93, N = 3 15846 15809
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: Triangle - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 90K 180K 270K 360K 450K SE +/- 300.42, N = 3 SE +/- 248.53, N = 3 409864 412804
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: Pixmark Piano - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 200 400 600 800 1000 1005 1000
OpenBenchmarking.org Points, More Is Better GpuTest 0.7.0 Test: Pixmark Volplosion - Resolution: 3840 x 2160 - Mode: Fullscreen GCC 9.2 Built Clang 9.0 Built 600 1200 1800 2400 3000 SE +/- 0.33, N = 3 2565 2549
Ethr Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Microseconds, Fewer Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1 GCC 9.2 Built Clang 9.0 Built 4 8 12 16 20 SE +/- 0.14, N = 15 SE +/- 0.21, N = 3 16.64 16.58 MIN: 6.1 / MAX: 22.77 MIN: 6.18 / MAX: 23.17
OpenBenchmarking.org Microseconds, Fewer Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 64 GCC 9.2 Built Clang 9.0 Built 4 8 12 16 20 SE +/- 0.09, N = 3 SE +/- 0.17, N = 15 16.45 16.55 MIN: 6.1 / MAX: 22.49 MIN: 6.09 / MAX: 22.82
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1 GCC 9.2 Built Clang 9.0 Built 200 400 600 800 1000 SE +/- 4.89, N = 5 SE +/- 4.45, N = 5 1005.77 1005.28 MAX: 1030 MAX: 1050
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 64 GCC 9.2 Built Clang 9.0 Built 40K 80K 120K 160K 200K SE +/- 619.26, N = 3 SE +/- 365.04, N = 3 157939.11 167530.67 MIN: 146710 / MAX: 169070 MIN: 163270 / MAX: 171280
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 64 GCC 9.2 Built Clang 9.0 Built 30K 60K 90K 120K 150K SE +/- 99.90, N = 3 SE +/- 246.51, N = 3 147127.78 144142.67 MIN: 140720 / MAX: 154930 MIN: 135600 / MAX: 150380
OpenBenchmarking.org Mbits/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 64 GCC 9.2 Built Clang 9.0 Built 800 1600 2400 3200 4000 SE +/- 6.97, N = 3 SE +/- 6.50, N = 3 3898.75 3924.17 MIN: 3800 / MAX: 4040 MIN: 3840 / MAX: 3990
OpenBenchmarking.org Connections/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1 GCC 9.2 Built Clang 9.0 Built 3K 6K 9K 12K 15K SE +/- 108.68, N = 3 SE +/- 186.32, N = 5 14643 14588
OpenBenchmarking.org Connections/sec, More Is Better Ethr 2019-01-02 Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 64 GCC 9.2 Built Clang 9.0 Built 12K 24K 36K 48K 60K SE +/- 3334.56, N = 15 SE +/- 7200.68, N = 15 51652 54385
NAS Parallel Benchmarks NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: BT.C GCC 9.2 Built Clang 9.0 Built 16K 32K 48K 64K 80K SE +/- 197.41, N = 3 SE +/- 382.78, N = 3 74188.70 74687.12 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C GCC 9.2 Built Clang 9.0 Built 2K 4K 6K 8K 10K SE +/- 27.38, N = 3 SE +/- 35.46, N = 3 11008.58 10998.94 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C GCC 9.2 Built Clang 9.0 Built 600 1200 1800 2400 3000 SE +/- 7.36, N = 3 SE +/- 8.17, N = 3 2923.71 2901.35 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D GCC 9.2 Built Clang 9.0 Built 600 1200 1800 2400 3000 SE +/- 4.31, N = 3 SE +/- 4.40, N = 3 2918.69 2923.07 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C GCC 9.2 Built Clang 9.0 Built 8K 16K 24K 32K 40K SE +/- 9.63, N = 3 SE +/- 100.06, N = 3 35126.47 34986.16 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D GCC 9.2 Built Clang 9.0 Built 300 600 900 1200 1500 SE +/- 6.71, N = 3 SE +/- 1.48, N = 3 1483.69 1491.41 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C GCC 9.2 Built Clang 9.0 Built 14K 28K 42K 56K 70K SE +/- 43.75, N = 3 SE +/- 96.45, N = 3 67155.42 66925.36 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C GCC 9.2 Built Clang 9.0 Built 7K 14K 21K 28K 35K SE +/- 23.30, N = 3 SE +/- 58.07, N = 3 30970.82 30893.36 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B GCC 9.2 Built Clang 9.0 Built 7K 14K 21K 28K 35K SE +/- 387.43, N = 3 SE +/- 123.63, N = 3 34456.00 34591.63 1. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi 2. Open MPI 3.1.3
OpenBenchmarking.org Nodes Per Second, More Is Better LeelaChessZero 0.22.0 Backend: Random GCC 9.2 Built Clang 9.0 Built 20K 40K 60K 80K 100K SE +/- 273.38, N = 3 SE +/- 366.88, N = 3 100269.0 100999.0 1. (CXX) g++ options: -lpthread
Parboil The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM GCC 9.2 Built Clang 9.0 Built 10 20 30 40 50 SE +/- 0.05, N = 3 SE +/- 0.00, N = 3 42.65 42.74 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP GCC 9.2 Built Clang 9.0 Built 0.2133 0.4266 0.6399 0.8532 1.0665 SE +/- 0.002941, N = 3 SE +/- 0.007491, N = 3 0.946099 0.948077 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil GCC 9.2 Built Clang 9.0 Built 1.0933 2.1866 3.2799 4.3732 5.4665 SE +/- 0.037115, N = 3 SE +/- 0.081857, N = 3 4.858992 4.782372 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding GCC 9.2 Built Clang 9.0 Built 13 26 39 52 65 SE +/- 0.07, N = 3 SE +/- 0.10, N = 3 58.27 58.27 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
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 the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD GCC 9.2 Built Clang 9.0 Built 2 4 6 8 10 SE +/- 0.012, N = 3 SE +/- 0.015, N = 3 8.008 8.024 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver GCC 9.2 Built Clang 9.0 Built 2 4 6 8 10 SE +/- 0.042, N = 3 SE +/- 0.016, N = 3 7.660 7.635 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster GCC 9.2 Built Clang 9.0 Built 4 8 12 16 20 SE +/- 0.23, N = 15 SE +/- 0.03, N = 3 14.96 14.62 1. (CXX) g++ options: -O2 -lOpenCL
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.
OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms GCC 9.2 Built Clang 9.0 Built 0.1269 0.2538 0.3807 0.5076 0.6345 SE +/- 0.00053, N = 12 SE +/- 0.00059, N = 3 0.56323 0.56379
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 0.8 Encoder Mode: Enc Mode 4 - Input: 1080p GCC 9.2 Built Clang 9.0 Built 2 4 6 8 10 SE +/- 0.026, N = 3 SE +/- 0.008, N = 3 8.430 8.347 1. (CXX) g++ options: -fPIE -fPIC -pie
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8 Encoder Mode: Enc Mode 8 - Input: 1080p GCC 9.2 Built Clang 9.0 Built 15 30 45 60 75 SE +/- 0.33, N = 3 SE +/- 0.19, N = 3 68.35 68.55 1. (CXX) g++ options: -fPIE -fPIC -pie
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 0.1 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p GCC 9.2 Built Clang 9.0 Built 80 160 240 320 400 SE +/- 4.28, N = 6 SE +/- 2.69, N = 13 351.32 353.11 1. (CC) gcc options: -fPIE -fPIC -fvisibility=hidden -O3 -pie -rdynamic -lpthread -lrt -lm
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 GCC 9.2 Built Clang 9.0 Built 20M 40M 60M 80M 100M SE +/- 760523.84, N = 3 SE +/- 1200577.83, N = 3 96431209 96599235 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto
DeepSpeech Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better DeepSpeech 0.6 Acceleration: CPU GCC 9.2 Built Clang 9.0 Built 13 26 39 52 65 SE +/- 0.26, N = 3 SE +/- 0.52, N = 3 56.05 57.87
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MT - Degridding GCC 9.2 Built Clang 9.0 Built 800 1600 2400 3200 4000 SE +/- 1.18, N = 3 SE +/- 3.13, N = 3 3882.23 3888.14 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MPI - Gridding GCC 9.2 Built Clang 9.0 Built 600 1200 1800 2400 3000 SE +/- 0.77, N = 3 SE +/- 1.63, N = 3 2734.92 2733.75 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve MPI - Degridding GCC 9.2 Built Clang 9.0 Built 800 1600 2400 3200 4000 SE +/- 0.60, N = 3 SE +/- 0.60, N = 3 3931.19 3927.57 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve OpenMP - Gridding GCC 9.2 Built Clang 9.0 Built 1400 2800 4200 5600 7000 SE +/- 0.00, N = 3 SE +/- 49.14, N = 3 6339.43 6290.29 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org Million Grid Points Per Second, More Is Better ASKAP 2018-11-10 Test: tConvolve OpenMP - Degridding GCC 9.2 Built Clang 9.0 Built 1000 2000 3000 4000 5000 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 4590.62 4590.62 1. (CXX) g++ options: -lpthread
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write GCC 9.2 Built Clang 9.0 Built 3K 6K 9K 12K 15K SE +/- 220.76, N = 15 SE +/- 1359.03, N = 15 12575.12 13491.88 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Only GCC 9.2 Built Clang 9.0 Built 160K 320K 480K 640K 800K SE +/- 4310.08, N = 3 SE +/- 2492.99, N = 3 758184.38 765627.63 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 12.0 Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write GCC 9.2 Built Clang 9.0 Built 3K 6K 9K 12K 15K SE +/- 202.28, N = 3 SE +/- 412.77, N = 15 11910.00 12792.71 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Memtier_benchmark Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool. This current test profile currently just stresses the Redis protocol and basic options exposed wotj a 1:1 Set/Get ratio, 30 pipeline, 100 clients per thread, and thread count equal to the number of CPU cores/threads present. Patches to extend the test are welcome as always. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ops/sec, More Is Better Memtier_benchmark 1.2.17 Protocol: Redis GCC 9.2 Built Clang 9.0 Built 700K 1400K 2100K 2800K 3500K SE +/- 87137.03, N = 15 SE +/- 103739.94, N = 12 3148364.33 3058231.03 1. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre
OpenBenchmarking.org Requests Per Second, More Is Better Redis 5.0.5 Test: SET GCC 9.2 Built Clang 9.0 Built 400K 800K 1200K 1600K 2000K SE +/- 27467.28, N = 4 SE +/- 29786.87, N = 15 1854760.03 1961726.78 1. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Semaphores GCC 9.2 Built Clang 9.0 Built 700K 1400K 2100K 2800K 3500K SE +/- 5354.29, N = 3 SE +/- 17095.02, N = 3 3378942.76 3378876.69 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Socket Activity GCC 9.2 Built Clang 9.0 Built 5K 10K 15K 20K 25K SE +/- 300.15, N = 3 SE +/- 107.95, N = 3 21782.55 22850.86 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Context Switching GCC 9.2 Built Clang 9.0 Built 3M 6M 9M 12M 15M SE +/- 70573.25, N = 3 SE +/- 32348.37, N = 3 12914746.02 13430897.91 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: System V Message Passing GCC 9.2 Built Clang 9.0 Built 4M 8M 12M 16M 20M SE +/- 56140.92, N = 3 SE +/- 80535.55, N = 3 19840008.52 16958863.51 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
Facebook RocksDB This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill GCC 9.2 Built Clang 9.0 Built 200K 400K 600K 800K 1000K SE +/- 2820.50, N = 3 SE +/- 2251.84, N = 3 1142437 1108679 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Read GCC 9.2 Built Clang 9.0 Built 40M 80M 120M 160M 200M SE +/- 416751.10, N = 3 SE +/- 1710859.72, N = 3 177770388 181240810 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Sequential Fill GCC 9.2 Built Clang 9.0 Built 300K 600K 900K 1200K 1500K SE +/- 700.57, N = 3 SE +/- 3104.32, N = 3 1211042 1173883 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Random Fill Sync GCC 9.2 Built Clang 9.0 Built 7K 14K 21K 28K 35K SE +/- 64.89, N = 3 SE +/- 132.59, N = 3 32546 32747 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.org Op/s, More Is Better Facebook RocksDB 6.3.6 Test: Read While Writing GCC 9.2 Built Clang 9.0 Built 1.3M 2.6M 3.9M 5.2M 6.5M SE +/- 28660.86, N = 3 SE +/- 47939.98, N = 3 6026743 6108992 1. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
Blender Blender is an open-source 3D creation software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL or CUDA is supported. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.81 Blend File: Pabellon Barcelona - Compute: CPU-Only GCC 9.2 Built Clang 9.0 Built 30 60 90 120 150 SE +/- 0.09, N = 3 SE +/- 0.39, N = 3 152.49 151.36
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 GCC 9.2 Built Clang 9.0 Built 9K 18K 27K 36K 45K SE +/- 637.10, N = 3 SE +/- 97.23, N = 3 42916.50 42955.88 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
Apache Benchmark This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better Apache Benchmark 2.4.29 Static Web Page Serving GCC 9.2 Built Clang 9.0 Built 7K 14K 21K 28K 35K SE +/- 78.79, N = 3 SE +/- 51.61, N = 3 34134.45 33679.09 1. (CC) gcc options: -shared -fPIC -O2 -pthread
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: WebXPRT - Browser: Firefox GCC 9.2 Built Clang 9.0 Built 60 120 180 240 300 SE +/- 0.33, N = 3 SE +/- 1.45, N = 3 261 259 1. firefox 71.0
OpenBenchmarking.org Overall Score, More Is Better Selenium Benchmark: Basemark - Browser: Firefox GCC 9.2 Built Clang 9.0 Built 160 320 480 640 800 SE +/- 10.68, N = 4 SE +/- 12.10, N = 12 741.21 754.81 1. firefox 71.0
OpenBenchmarking.org Score, More Is Better Selenium Benchmark: CanvasMark - Browser: Firefox GCC 9.2 Built Clang 9.0 Built 3K 6K 9K 12K 15K SE +/- 207.75, N = 12 SE +/- 131.56, N = 12 14907 14887 1. firefox 71.0
OpenBenchmarking.org Score, Fewer Is Better Selenium Benchmark: PSPDFKit WASM - Browser: Firefox GCC 9.2 Built Clang 9.0 Built 300 600 900 1200 1500 SE +/- 4.70, N = 3 SE +/- 2.85, N = 3 1304 1297 1. firefox 71.0
Clang 9.0 Built Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0601 BIOS), Chipset: AMD Starship/Matisse, Memory: 64512MB, Disk: 1000GB Force MP600, Graphics: AMD Radeon VII 16GB (1802/1001MHz), Audio: AMD Vega 20 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 19.10, Kernel: 5.5.0-rc3-clang9phx (x86_64) 20191224, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: amdgpu 19.0.1, OpenGL: 4.5 Mesa 19.2.1 (LLVM 9.0.0), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 3840x2160
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Graphics Notes: GLAMORPython Notes: Python 2.7.17rc1 + Python 3.7.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected
Testing initiated at 24 December 2019 10:13 by user pts.
GCC 9.2 Built Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0601 BIOS), Chipset: AMD Starship/Matisse, Memory: 64512MB, Disk: 1000GB Force MP600, Graphics: AMD Radeon VII 16GB (1802/1001MHz), Audio: AMD Vega 20 HDMI Audio, Monitor: ASUS VP28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200
OS: Ubuntu 19.10, Kernel: 5.5.0-rc3-gcc9phx (x86_64) 20191224, Desktop: GNOME Shell 3.34.1, Display Server: X Server 1.20.5, Display Driver: amdgpu 19.0.1, OpenGL: 4.5 Mesa 19.2.1 (LLVM 9.0.0), Compiler: GCC 9.2.1 20191008, File-System: ext4, Screen Resolution: 3840x2160
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-offload-targets=nvptx-none,hsa --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-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 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025Graphics Notes: GLAMORPython Notes: Python 2.7.17rc1 + Python 3.7.5Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected
Testing initiated at 25 December 2019 05:16 by user pts.