2 x AMD EPYC 75F3 32-Core testing with a ASRockRack ROME2D16-2T (P3.10 BIOS) and ASPEED on Ubuntu 21.10 via the Phoronix Test Suite.
Linux 5.14 Processor: 2 x AMD EPYC 75F3 32-Core @ 2.95GHz (64 Cores / 128 Threads), Motherboard: ASRockRack ROME2D16-2T (P3.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel 10G X550T
OS: Ubuntu 21.10, Kernel: 5.14.0-051400-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server, Vulkan: 1.1.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001114Python Notes: Python 3.9.7Security 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 IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Linux 5.15 OS: Ubuntu 21.10, Kernel: 5.15.0-051500-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server, Vulkan: 1.1.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
Linux 5.16 Git OS: Ubuntu 21.10, Kernel: 5.16.0-051600rc2daily20211124-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server, Vulkan: 1.1.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001114Python Notes: Python 3.9.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Flexible IO Tester FIO, the Flexible I/O Tester, is an advanced Linux disk benchmark supporting multiple I/O engines and a wealth of options. FIO was written by Jens Axboe for testing of the Linux I/O subsystem and schedulers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.14 Linux 5.16 Git Linux 5.15 1400 2800 4200 5600 7000 SE +/- 92.52, N = 3 SE +/- 26.85, N = 3 SE +/- 34.12, N = 3 6535 6548 6629 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.14 Linux 5.16 Git Linux 5.15 700 1400 2100 2800 3500 SE +/- 46.23, N = 3 SE +/- 13.35, N = 3 SE +/- 17.06, N = 3 3264 3270 3311 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.16 Git Linux 5.15 Linux 5.14 300 600 900 1200 1500 SE +/- 13.61, N = 3 SE +/- 9.53, N = 3 SE +/- 16.93, N = 4 1209 1276 1429 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.16 Git Linux 5.15 Linux 5.14 80K 160K 240K 320K 400K SE +/- 3527.67, N = 3 SE +/- 2403.70, N = 3 SE +/- 4415.88, N = 4 309333 326333 366000 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 1000 2000 3000 4000 5000 SE +/- 3.38, N = 3 SE +/- 48.29, N = 3 SE +/- 5.29, N = 3 4675 4681 4819 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 500 1000 1500 2000 2500 SE +/- 1.86, N = 3 SE +/- 24.04, N = 3 SE +/- 2.65, N = 3 2334 2337 2406 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 300 600 900 1200 1500 SE +/- 6.68, N = 12 SE +/- 14.76, N = 15 SE +/- 10.41, N = 15 941 1005 1331 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 70K 140K 210K 280K 350K SE +/- 1683.25, N = 12 SE +/- 3781.91, N = 15 SE +/- 2620.07, N = 15 241000 257400 340600 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 1400 2800 4200 5600 7000 SE +/- 72.43, N = 5 SE +/- 71.37, N = 3 SE +/- 5.51, N = 3 6577 6622 6719 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 700 1400 2100 2800 3500 SE +/- 36.27, N = 5 SE +/- 35.69, N = 3 SE +/- 2.65, N = 3 3285 3307 3356 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.16 Git Linux 5.15 Linux 5.14 300 600 900 1200 1500 SE +/- 15.06, N = 3 SE +/- 16.53, N = 15 SE +/- 2.73, N = 3 1047 1059 1465 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.16 Git Linux 5.15 Linux 5.14 80K 160K 240K 320K 400K SE +/- 4041.45, N = 3 SE +/- 4247.65, N = 15 SE +/- 666.67, N = 3 268000 271067 374667 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 1100 2200 3300 4400 5500 SE +/- 10.60, N = 3 SE +/- 3.18, N = 3 SE +/- 42.46, N = 3 4868 4869 4948 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 500 1000 1500 2000 2500 SE +/- 5.24, N = 3 SE +/- 1.53, N = 3 SE +/- 21.37, N = 3 2430 2431 2470 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 300 600 900 1200 1500 SE +/- 5.84, N = 3 SE +/- 5.04, N = 3 SE +/- 10.97, N = 9 924 1030 1317 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org IOPS, More Is Better Flexible IO Tester 3.25 Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test Directory Linux 5.15 Linux 5.16 Git Linux 5.14 70K 140K 210K 280K 350K SE +/- 1666.67, N = 3 SE +/- 1201.85, N = 3 SE +/- 2788.87, N = 9 236333 263333 337000 1. (CC) gcc options: -rdynamic -ll -lnuma -lz -lm -lpthread -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 5000 Files, 1MB Size, 4 Threads Linux 5.16 Git Linux 5.14 Linux 5.15 300 600 900 1200 1500 SE +/- 31.19, N = 15 SE +/- 72.17, N = 12 SE +/- 37.42, N = 15 1136.9 1216.6 1423.7
OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size Linux 5.14 Linux 5.16 Git Linux 5.15 160 320 480 640 800 SE +/- 23.95, N = 12 SE +/- 27.49, N = 12 SE +/- 22.00, N = 15 551.3 635.5 725.6
OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size, No Sync/FSync Linux 5.14 Linux 5.15 Linux 5.16 Git 600 1200 1800 2400 3000 SE +/- 5.77, N = 3 SE +/- 18.24, N = 12 SE +/- 17.76, N = 3 2312.6 2556.5 2632.4
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 Linux 5.14 Linux 5.16 Git Linux 5.15 500 1000 1500 2000 2500 SE +/- 11.18, N = 3 SE +/- 4.52, N = 3 SE +/- 2.61, N = 3 2227.86 2331.37 2361.17
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create Linux 5.16 Git Linux 5.14 Linux 5.15 130 260 390 520 650 SE +/- 1.39, N = 3 SE +/- 2.25, N = 3 SE +/- 5.62, N = 3 584.97 586.47 610.69
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Read Compiled Tree Linux 5.14 Linux 5.16 Git Linux 5.15 800 1600 2400 3200 4000 SE +/- 45.77, N = 3 SE +/- 74.42, N = 3 SE +/- 44.29, N = 3 3255.15 3544.96 3571.60
Sockperf This is a network socket API performance benchmark developed by Mellanox. This test profile runs both the client and server on the local host for evaluating individual system performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.7 Test: Throughput Linux 5.14 Linux 5.15 Linux 5.16 Git 150K 300K 450K 600K 750K SE +/- 4176.53, N = 5 SE +/- 5508.05, N = 5 SE +/- 3286.41, N = 25 693327 696621 697239 1. (CXX) g++ options: --param -O3 -rdynamic
OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.7 Test: Latency Ping Pong Linux 5.14 Linux 5.15 Linux 5.16 Git 1.0607 2.1214 3.1821 4.2428 5.3035 SE +/- 0.036, N = 5 SE +/- 0.179, N = 25 SE +/- 0.207, N = 25 4.714 4.402 4.301 1. (CXX) g++ options: --param -O3 -rdynamic
OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.7 Test: Latency Under Load Linux 5.14 Linux 5.16 Git Linux 5.15 3 6 9 12 15 SE +/- 0.17, N = 24 SE +/- 0.11, N = 7 SE +/- 0.11, N = 25 12.48 12.38 12.35 1. (CXX) g++ options: --param -O3 -rdynamic
OSBench OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Files Linux 5.15 Linux 5.16 Git Linux 5.14 4 8 12 16 20 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 0.10, N = 3 14.55 14.50 14.12 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Threads Linux 5.16 Git Linux 5.14 Linux 5.15 5 10 15 20 25 SE +/- 0.18, N = 3 SE +/- 0.12, N = 3 SE +/- 0.16, N = 3 22.63 22.14 21.35 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Launch Programs Linux 5.16 Git Linux 5.15 Linux 5.14 13 26 39 52 65 SE +/- 0.57, N = 3 SE +/- 0.63, N = 3 SE +/- 0.07, N = 3 57.25 57.04 54.91 1. (CC) gcc options: -lm
OpenBenchmarking.org us Per Event, Fewer Is Better OSBench Test: Create Processes Linux 5.16 Git Linux 5.15 Linux 5.14 10 20 30 40 50 SE +/- 0.58, N = 3 SE +/- 0.15, N = 3 SE +/- 0.20, N = 3 45.72 44.25 44.18 1. (CC) gcc options: -lm
OpenBenchmarking.org Ns Per Event, Fewer Is Better OSBench Test: Memory Allocations Linux 5.16 Git Linux 5.14 Linux 5.15 15 30 45 60 75 SE +/- 0.44, N = 3 SE +/- 0.39, N = 3 SE +/- 0.01, N = 3 67.17 66.43 66.17 1. (CC) gcc options: -lm
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 Linux 5.15 Linux 5.16 Git Linux 5.14 40K 80K 120K 160K 200K SE +/- 1310.81, N = 3 SE +/- 732.12, N = 3 SE +/- 84.87, N = 3 187436.39 187464.07 187950.73 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: CG.C Linux 5.15 Linux 5.16 Git Linux 5.14 9K 18K 27K 36K 45K SE +/- 630.56, N = 15 SE +/- 587.97, N = 12 SE +/- 424.43, N = 15 37683.20 38505.09 40391.37 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.C Linux 5.16 Git Linux 5.14 Linux 5.15 1300 2600 3900 5200 6500 SE +/- 41.13, N = 3 SE +/- 45.99, N = 10 SE +/- 50.98, N = 3 5892.73 6019.68 6034.63 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: EP.D Linux 5.16 Git Linux 5.15 Linux 5.14 1300 2600 3900 5200 6500 SE +/- 81.54, N = 3 SE +/- 76.36, N = 12 SE +/- 64.62, N = 15 6230.39 6267.76 6268.38 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: FT.C Linux 5.14 Linux 5.15 Linux 5.16 Git 20K 40K 60K 80K 100K SE +/- 962.01, N = 4 SE +/- 295.62, N = 3 SE +/- 200.12, N = 3 83950.22 85857.53 85869.54 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: IS.D Linux 5.16 Git Linux 5.14 Linux 5.15 800 1600 2400 3200 4000 SE +/- 28.48, N = 3 SE +/- 13.29, N = 3 SE +/- 24.72, N = 3 3517.61 3521.88 3587.80 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: LU.C Linux 5.14 Linux 5.16 Git Linux 5.15 60K 120K 180K 240K 300K SE +/- 1476.90, N = 3 SE +/- 2354.59, N = 3 SE +/- 1536.62, N = 3 257715.41 258932.20 260133.03 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: MG.C Linux 5.14 Linux 5.16 Git Linux 5.15 20K 40K 60K 80K 100K SE +/- 319.61, N = 3 SE +/- 568.12, N = 3 SE +/- 369.92, N = 3 80504.43 81198.74 81255.47 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.B Linux 5.15 Linux 5.14 Linux 5.16 Git 30K 60K 90K 120K 150K SE +/- 1294.01, N = 9 SE +/- 596.50, N = 3 SE +/- 369.75, N = 3 155640.09 159135.95 159697.08 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
OpenBenchmarking.org Total Mop/s, More Is Better NAS Parallel Benchmarks 3.4 Test / Class: SP.C Linux 5.14 Linux 5.16 Git Linux 5.15 20K 40K 60K 80K 100K SE +/- 95.57, N = 3 SE +/- 201.10, N = 3 SE +/- 98.59, N = 3 83630.80 83935.50 84059.02 1. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.0
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 Linux 5.14 Linux 5.15 Linux 5.16 Git 7 14 21 28 35 SE +/- 0.87, N = 12 SE +/- 1.17, N = 12 SE +/- 0.49, N = 15 31.65 31.35 31.30 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP Linux 5.16 Git Linux 5.14 Linux 5.15 0.1979 0.3958 0.5937 0.7916 0.9895 SE +/- 0.005811, N = 3 SE +/- 0.007978, N = 3 SE +/- 0.005555, N = 3 0.879649 0.870178 0.868256 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Test: OpenMP MRI-Q
Linux 5.14: main.c:(.text.startup+0x20c): undefined reference to `ComputeQCPU'
Linux 5.15: main.c:(.text.startup+0x20c): undefined reference to `ComputeQCPU'
Linux 5.16 Git: main.c:(.text.startup+0x20c): undefined reference to `ComputeQCPU'
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP Stencil Linux 5.15 Linux 5.14 Linux 5.16 Git 1.1783 2.3566 3.5349 4.7132 5.8915 SE +/- 0.063237, N = 3 SE +/- 0.009600, N = 3 SE +/- 0.047938, N = 14 5.236831 5.137429 5.122243 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP MRI Gridding Linux 5.15 Linux 5.16 Git Linux 5.14 60 120 180 240 300 SE +/- 2.18, N = 3 SE +/- 1.04, N = 3 SE +/- 2.83, N = 3 286.76 285.68 283.59 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 select OpenCL, NVIDIA CUDA and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP LavaMD Linux 5.14 Linux 5.15 Linux 5.16 Git 8 16 24 32 40 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.11, N = 3 36.81 36.42 36.39 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP HotSpot3D Linux 5.15 Linux 5.14 Linux 5.16 Git 20 40 60 80 100 SE +/- 1.13, N = 15 SE +/- 1.47, N = 12 SE +/- 0.12, N = 3 78.88 77.79 73.89 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Leukocyte Linux 5.15 Linux 5.14 Linux 5.16 Git 10 20 30 40 50 SE +/- 0.33, N = 3 SE +/- 0.31, N = 12 SE +/- 0.54, N = 4 44.77 44.41 43.86 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP CFD Solver Linux 5.16 Git Linux 5.14 Linux 5.15 2 4 6 8 10 SE +/- 0.057, N = 3 SE +/- 0.009, N = 3 SE +/- 0.050, N = 15 7.049 6.988 6.877 1. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 3.1 Test: OpenMP Streamcluster Linux 5.14 Linux 5.16 Git Linux 5.15 2 4 6 8 10 SE +/- 0.357, N = 12 SE +/- 0.098, N = 15 SE +/- 0.134, N = 15 7.991 7.612 7.578 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 2.14 ATPase Simulation - 327,506 Atoms Linux 5.14 Linux 5.16 Git Linux 5.15 0.0686 0.1372 0.2058 0.2744 0.343 SE +/- 0.00078, N = 3 SE +/- 0.00003, N = 3 SE +/- 0.00017, N = 3 0.30492 0.30451 0.30429
Zstd Compression This test measures the time needed to compress/decompress a sample input file using Zstd compression supplied by the system or otherwise externally of the test profile. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 8 - Compression Speed Linux 5.16 Git Linux 5.14 Linux 5.15 700 1400 2100 2800 3500 SE +/- 53.95, N = 15 SE +/- 45.90, N = 15 SE +/- 33.74, N = 15 3176.6 3241.8 3262.9 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 8 - Decompression Speed Linux 5.16 Git Linux 5.14 Linux 5.15 800 1600 2400 3200 4000 SE +/- 29.53, N = 4 SE +/- 29.97, N = 3 SE +/- 20.28, N = 3 3861.4 3891.5 3912.5 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 19 - Compression Speed Linux 5.15 Linux 5.14 Linux 5.16 Git 20 40 60 80 100 SE +/- 0.74, N = 6 SE +/- 0.69, N = 15 SE +/- 0.91, N = 3 76.7 78.1 78.6 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 19 - Decompression Speed Linux 5.14 Linux 5.16 Git Linux 5.15 800 1600 2400 3200 4000 SE +/- 13.40, N = 15 SE +/- 15.19, N = 3 SE +/- 15.35, N = 6 3538.9 3580.2 3607.0 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 8, Long Mode - Compression Speed Linux 5.14 Linux 5.16 Git Linux 5.15 40 80 120 160 200 SE +/- 2.47, N = 15 SE +/- 2.50, N = 3 SE +/- 5.22, N = 15 186.0 193.1 196.2 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 8, Long Mode - Decompression Speed Linux 5.16 Git Linux 5.14 Linux 5.15 900 1800 2700 3600 4500 SE +/- 125.76, N = 3 SE +/- 14.05, N = 15 SE +/- 16.94, N = 15 4094.5 4115.7 4117.1 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 19, Long Mode - Compression Speed Linux 5.14 Linux 5.16 Git Linux 5.15 9 18 27 36 45 SE +/- 0.39, N = 15 SE +/- 0.67, N = 12 SE +/- 0.26, N = 3 39.0 39.1 39.4 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
OpenBenchmarking.org MB/s, More Is Better Zstd Compression Compression Level: 19, Long Mode - Decompression Speed Linux 5.14 Linux 5.15 Linux 5.16 Git 800 1600 2400 3200 4000 SE +/- 6.50, N = 15 SE +/- 44.80, N = 3 SE +/- 11.44, N = 12 3599.8 3608.5 3609.9 1. *** zstd command line interface 64-bits v1.4.8, by Yann Collet ***
JPEG XL libjxl The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is currently focused on the multi-threaded JPEG XL image encode performance using the reference libjxl library. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.6.1 Input: PNG - Encode Speed: 7 Linux 5.14 Linux 5.16 Git Linux 5.15 3 6 9 12 15 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 9.40 9.44 9.49 1. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.6.1 Input: PNG - Encode Speed: 8 Linux 5.14 Linux 5.15 Linux 5.16 Git 0.216 0.432 0.648 0.864 1.08 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 0.95 0.96 0.96 1. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.6.1 Input: JPEG - Encode Speed: 7 Linux 5.15 Linux 5.16 Git Linux 5.14 16 32 48 64 80 SE +/- 0.52, N = 3 SE +/- 0.14, N = 3 SE +/- 0.43, N = 3 69.57 70.00 70.37 1. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
OpenBenchmarking.org MP/s, More Is Better JPEG XL libjxl 0.6.1 Input: JPEG - Encode Speed: 8 Linux 5.15 Linux 5.14 Linux 5.16 Git 7 14 21 28 35 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 SE +/- 0.19, N = 3 27.92 28.15 28.35 1. (CXX) g++ options: -funwind-tables -O3 -O2 -fPIE -pie
JPEG XL Decoding libjxl The JPEG XL Image Coding System is designed to provide next-generation JPEG image capabilities with JPEG XL offering better image quality and compression over legacy JPEG. This test profile is suited for JPEG XL decode performance testing to PNG output file, the pts/jpexl test is for encode performance. The JPEG XL encoding/decoding is done using the libjxl codebase. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MP/s, More Is Better JPEG XL Decoding libjxl 0.6.1 CPU Threads: 1 Linux 5.14 Linux 5.16 Git Linux 5.15 13 26 39 52 65 SE +/- 0.10, N = 3 SE +/- 0.01, N = 3 SE +/- 0.09, N = 3 54.84 56.19 57.56
AOM AV1 This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.2 Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4K Linux 5.15 Linux 5.14 Linux 5.16 Git 1.332 2.664 3.996 5.328 6.66 SE +/- 0.08, N = 3 SE +/- 0.02, N = 3 SE +/- 0.04, N = 3 5.86 5.89 5.92 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.2 Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4K Linux 5.14 Linux 5.15 Linux 5.16 Git 2 4 6 8 10 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 8.23 8.38 8.43 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.2 Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4K Linux 5.16 Git Linux 5.15 Linux 5.14 6 12 18 24 30 SE +/- 0.26, N = 3 SE +/- 0.06, N = 3 SE +/- 0.32, N = 4 25.88 26.05 26.78 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.2 Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4K Linux 5.14 Linux 5.16 Git Linux 5.15 8 16 24 32 40 SE +/- 0.41, N = 15 SE +/- 0.26, N = 15 SE +/- 0.14, N = 3 35.75 35.89 36.31 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
OpenBenchmarking.org Frames Per Second, More Is Better AOM AV1 3.2 Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4K Linux 5.15 Linux 5.16 Git Linux 5.14 9 18 27 36 45 SE +/- 0.33, N = 3 SE +/- 0.30, N = 3 SE +/- 0.02, N = 3 39.66 40.16 40.32 1. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm
Kvazaar This is a test of Kvazaar as a CPU-based H.265/HEVC video encoder written in the C programming language and optimized in Assembly. Kvazaar is the winner of the 2016 ACM Open-Source Software Competition and developed at the Ultra Video Group, Tampere University, Finland. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Slow Linux 5.14 Linux 5.16 Git Linux 5.15 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.03, N = 3 14.70 14.79 14.97 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Medium Linux 5.15 Linux 5.14 Linux 5.16 Git 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.01, N = 3 14.90 14.91 14.97 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 1080p - Video Preset: Slow Linux 5.16 Git Linux 5.15 Linux 5.14 12 24 36 48 60 SE +/- 0.13, N = 3 SE +/- 0.75, N = 3 SE +/- 0.77, N = 3 54.41 54.67 55.60 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 1080p - Video Preset: Medium Linux 5.15 Linux 5.16 Git Linux 5.14 13 26 39 52 65 SE +/- 0.68, N = 3 SE +/- 0.18, N = 3 SE +/- 0.40, N = 15 54.95 55.02 58.36 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Very Fast Linux 5.14 Linux 5.16 Git Linux 5.15 6 12 18 24 30 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.18, N = 3 25.70 25.81 26.27 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 4K - Video Preset: Ultra Fast Linux 5.14 Linux 5.15 Linux 5.16 Git 11 22 33 44 55 SE +/- 0.38, N = 3 SE +/- 0.21, N = 3 SE +/- 0.35, N = 3 49.21 49.56 49.69 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 1080p - Video Preset: Very Fast Linux 5.14 Linux 5.16 Git Linux 5.15 30 60 90 120 150 SE +/- 1.35, N = 15 SE +/- 1.28, N = 15 SE +/- 0.84, N = 3 125.14 131.77 135.89 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better Kvazaar 2.1 Video Input: Bosphorus 1080p - Video Preset: Ultra Fast Linux 5.14 Linux 5.16 Git Linux 5.15 50 100 150 200 250 SE +/- 0.93, N = 3 SE +/- 1.57, N = 3 SE +/- 2.11, N = 15 235.97 239.24 244.53 1. (CC) gcc options: -pthread -ftree-vectorize -fvisibility=hidden -O2 -lpthread -lm -lrt
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.5 Speed: 5 Linux 5.14 Linux 5.15 Linux 5.16 Git 0.5164 1.0328 1.5492 2.0656 2.582 SE +/- 0.009, N = 3 SE +/- 0.024, N = 4 SE +/- 0.011, N = 3 2.262 2.285 2.295
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.5 Speed: 6 Linux 5.16 Git Linux 5.14 Linux 5.15 0.6575 1.315 1.9725 2.63 3.2875 SE +/- 0.020, N = 3 SE +/- 0.022, N = 11 SE +/- 0.031, N = 3 2.913 2.916 2.922
OpenBenchmarking.org Frames Per Second, More Is Better rav1e 0.5 Speed: 10 Linux 5.14 Linux 5.16 Git Linux 5.15 2 4 6 8 10 SE +/- 0.050, N = 15 SE +/- 0.063, N = 3 SE +/- 0.031, N = 3 6.343 6.367 6.377
SVT-AV1 This is a benchmark of the SVT-AV1 open-source video encoder/decoder. SVT-AV1 was originally developed by Intel as part of their Open Visual Cloud / Scalable Video Technology (SVT). Development of SVT-AV1 has since moved to the Alliance for Open Media as part of upstream AV1 development. SVT-AV1 is a CPU-based multi-threaded video encoder for the AV1 video format with a sample YUV video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 4 - Input: Bosphorus 4K Linux 5.14 Linux 5.15 Linux 5.16 Git 1.2645 2.529 3.7935 5.058 6.3225 SE +/- 0.038, N = 3 SE +/- 0.024, N = 3 SE +/- 0.010, N = 3 5.552 5.585 5.620 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
OpenBenchmarking.org Frames Per Second, More Is Better SVT-AV1 0.8.7 Encoder Mode: Preset 8 - Input: Bosphorus 4K Linux 5.14 Linux 5.16 Git Linux 5.15 13 26 39 52 65 SE +/- 0.66, N = 4 SE +/- 0.67, N = 3 SE +/- 0.41, N = 3 58.73 59.07 60.24 1. (CXX) g++ options: -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq -pie
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 1.5.0 Tuning: 1 - Input: Bosphorus 1080p Linux 5.14 Linux 5.16 Git Linux 5.15 11 22 33 44 55 SE +/- 0.31, N = 3 SE +/- 0.25, N = 3 SE +/- 0.27, N = 3 48.36 48.84 48.90 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 7 - Input: Bosphorus 1080p Linux 5.14 Linux 5.16 Git Linux 5.15 80 160 240 320 400 SE +/- 0.56, N = 3 SE +/- 2.50, N = 3 SE +/- 3.88, N = 3 343.71 349.62 350.49 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt
OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 1.5.0 Tuning: 10 - Input: Bosphorus 1080p Linux 5.16 Git Linux 5.14 Linux 5.15 120 240 360 480 600 SE +/- 2.30, N = 3 SE +/- 4.95, N = 3 SE +/- 7.77, N = 3 523.12 527.64 543.37 1. (CC) gcc options: -fPIE -fPIC -O3 -O2 -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 YUV input video file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: VMAF Optimized - Input: Bosphorus 1080p Linux 5.14 Linux 5.16 Git Linux 5.15 80 160 240 320 400 SE +/- 14.65, N = 12 SE +/- 16.77, N = 12 SE +/- 6.38, N = 15 375.18 389.01 389.96 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080p Linux 5.15 Linux 5.14 Linux 5.16 Git 80 160 240 320 400 SE +/- 4.71, N = 15 SE +/- 4.71, N = 15 SE +/- 4.25, N = 15 379.00 387.12 389.40 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 0.3 Tuning: Visual Quality Optimized - Input: Bosphorus 1080p Linux 5.14 Linux 5.16 Git Linux 5.15 70 140 210 280 350 SE +/- 2.75, N = 3 SE +/- 2.45, N = 3 SE +/- 2.59, N = 3 308.76 320.81 330.81 1. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm
Stargate Digital Audio Workstation Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 21.10.9 Sample Rate: 44100 - Buffer Size: 512 Linux 5.14 1.2732 2.5464 3.8196 5.0928 6.366 SE +/- 0.028941, N = 3 5.658679 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 21.10.9 Sample Rate: 96000 - Buffer Size: 512 Linux 5.14 0.9083 1.8166 2.7249 3.6332 4.5415 SE +/- 0.023685, N = 3 4.036974 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
OpenBenchmarking.org Render Ratio, More Is Better Stargate Digital Audio Workstation 21.10.9 Sample Rate: 192000 - Buffer Size: 512 Linux 5.14 0.5846 1.1692 1.7538 2.3384 2.923 SE +/- 0.039181, N = 2 2.598303 1. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions
Sample Rate: 44100 - Buffer Size: 1024
Linux 5.14: Test failed to run.
Sample Rate: 480000 - Buffer Size: 512
Linux 5.14: Test failed to run.
Sample Rate: 96000 - Buffer Size: 1024
Linux 5.14: Test failed to run.
Sample Rate: 192000 - Buffer Size: 1024
Linux 5.14: Test failed to run.
Sample Rate: 480000 - Buffer Size: 1024
Linux 5.14: Test failed to run.
Timed Wasmer Compilation This test times how long it takes to compile Wasmer. Wasmer is written in the Rust programming language and is a WebAssembly runtime implementation that supports WASI and EmScripten. This test profile builds Wasmer with the Cranelift and Singlepast compiler features enabled. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Timed Wasmer Compilation 1.0.2 Time To Compile Linux 5.15 Linux 5.16 Git Linux 5.14 8 16 24 32 40 SE +/- 0.08, N = 3 SE +/- 0.08, N = 3 SE +/- 0.11, N = 3 35.03 34.82 34.67 1. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lgcc_s -lutil -lrt -lpthread -lm -lc
Google SynthMark SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Voices, More Is Better Google SynthMark 20201109 Test: VoiceMark_100 Linux 5.15 Linux 5.14 Linux 5.16 Git 200 400 600 800 1000 SE +/- 4.61, N = 3 SE +/- 1.23, N = 3 SE +/- 0.48, N = 3 808.41 815.44 816.22 1. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
Cpuminer-Opt Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Magi Linux 5.16 Git Linux 5.14 Linux 5.15 600 1200 1800 2400 3000 SE +/- 157.98, N = 15 SE +/- 144.19, N = 15 SE +/- 119.46, N = 15 2471.90 2499.95 2673.13 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: x25x Linux 5.14 Linux 5.15 Linux 5.16 Git 400 800 1200 1600 2000 SE +/- 27.19, N = 3 SE +/- 21.63, N = 15 SE +/- 33.22, N = 15 1885.00 1928.94 2025.73 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Deepcoin Linux 5.16 Git Linux 5.14 Linux 5.15 5K 10K 15K 20K 25K SE +/- 1084.56, N = 12 SE +/- 3672.19, N = 15 SE +/- 5013.61, N = 15 16143.00 20219.74 24318.96 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Ringcoin Linux 5.15 Linux 5.14 Linux 5.16 Git 200 400 600 800 1000 SE +/- 213.31, N = 12 SE +/- 239.72, N = 15 SE +/- 239.77, N = 15 505.10 985.76 1147.29 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Blake-2 S Linux 5.14 Linux 5.16 Git Linux 5.15 200K 400K 600K 800K 1000K SE +/- 11078.02, N = 3 SE +/- 8147.41, N = 3 SE +/- 52176.46, N = 12 954850 961537 1012175 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Garlicoin Linux 5.15 Linux 5.14 Linux 5.16 Git 300 600 900 1200 1500 SE +/- 12.33, N = 15 SE +/- 12.92, N = 4 SE +/- 16.53, N = 15 1149.48 1165.34 1188.38 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Skeincoin Linux 5.16 Git Linux 5.15 Linux 5.14 60K 120K 180K 240K 300K SE +/- 1597.16, N = 3 SE +/- 1049.63, N = 3 SE +/- 1778.18, N = 15 252150 255050 259079 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Myriad-Groestl Linux 5.15 Linux 5.14 Linux 5.16 Git 30K 60K 90K 120K 150K SE +/- 11547.72, N = 15 SE +/- 12143.76, N = 12 SE +/- 10162.55, N = 12 108893.00 112178.00 120101.05 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: LBC, LBRY Credits Linux 5.15 Linux 5.14 Linux 5.16 Git 20K 40K 60K 80K 100K SE +/- 1318.65, N = 3 SE +/- 56.96, N = 3 SE +/- 1030.04, N = 7 111503 112877 113431 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Quad SHA-256, Pyrite Linux 5.16 Git Linux 5.14 Linux 5.15 30K 60K 90K 120K 150K SE +/- 416.67, N = 3 SE +/- 166.67, N = 3 SE +/- 220.48, N = 3 127167 127333 127583 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.org kH/s, More Is Better Cpuminer-Opt 3.18 Algorithm: Triple SHA-256, Onecoin Linux 5.15 Linux 5.14 Linux 5.16 Git 50K 100K 150K 200K 250K SE +/- 294.18, N = 3 SE +/- 600.93, N = 3 SE +/- 333.33, N = 3 254443 254833 255667 1. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. The system/openssl test profiles relies on benchmarking the system/OS-supplied openssl binary rather than the pts/openssl test profile that uses the locally-built OpenSSL for benchmarking. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org sign/s, More Is Better OpenSSL Linux 5.16 Git Linux 5.14 Linux 5.15 4K 8K 12K 16K 20K SE +/- 9.75, N = 3 SE +/- 12.69, N = 3 SE +/- 2.34, N = 3 17090.7 17097.3 17109.8 1. OpenSSL 1.1.1l 24 Aug 2021
OpenBenchmarking.org verify/s, More Is Better OpenSSL Linux 5.15 Linux 5.14 Linux 5.16 Git 200K 400K 600K 800K 1000K SE +/- 198.67, N = 3 SE +/- 295.42, N = 3 SE +/- 266.29, N = 3 1101646.3 1103127.8 1104766.1 1. OpenSSL 1.1.1l 24 Aug 2021
KeyDB A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ops/sec, More Is Better KeyDB 6.2.0 Linux 5.16 Git Linux 5.15 Linux 5.14 20K 40K 60K 80K 100K SE +/- 477.02, N = 15 SE +/- 2817.21, N = 15 SE +/- 3029.01, N = 13 67689.41 88643.42 99646.77 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
GROMACS The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Ns Per Day, More Is Better GROMACS 2021.2 Implementation: MPI CPU - Input: water_GMX50_bare Linux 5.16 Git Linux 5.15 Linux 5.14 2 4 6 8 10 SE +/- 0.139, N = 15 SE +/- 0.076, N = 3 SE +/- 0.011, N = 3 7.172 7.342 7.381 1. (CXX) g++ options: -O3
PostgreSQL pgbench This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only Linux 5.15 Linux 5.14 Linux 5.16 Git 400K 800K 1200K 1600K 2000K SE +/- 4184.99, N = 3 SE +/- 8367.18, N = 3 SE +/- 11831.17, N = 3 1662624 1679150 1726379 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average Latency Linux 5.15 Linux 5.14 Linux 5.16 Git 0.0338 0.0676 0.1014 0.1352 0.169 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.150 0.149 0.145 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 500 - Mode: Read Only Linux 5.16 Git Linux 5.14 Linux 5.15 400K 800K 1200K 1600K 2000K SE +/- 5469.42, N = 3 SE +/- 16032.30, N = 3 SE +/- 23255.77, N = 3 1839066 1844914 1863564 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 500 - Mode: Read Only - Average Latency Linux 5.16 Git Linux 5.14 Linux 5.15 0.0612 0.1224 0.1836 0.2448 0.306 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 SE +/- 0.003, N = 3 0.272 0.271 0.268 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write Linux 5.16 Git Linux 5.14 Linux 5.15 9K 18K 27K 36K 45K SE +/- 561.10, N = 3 SE +/- 808.82, N = 12 SE +/- 1689.89, N = 12 41856 43578 44152 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average Latency Linux 5.16 Git Linux 5.15 Linux 5.14 1.3444 2.6888 4.0332 5.3776 6.722 SE +/- 0.081, N = 3 SE +/- 0.318, N = 12 SE +/- 0.105, N = 12 5.975 5.794 5.758 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 500 - Mode: Read Write Linux 5.16 Git Linux 5.14 Linux 5.15 10K 20K 30K 40K 50K SE +/- 2646.26, N = 9 SE +/- 360.43, N = 9 SE +/- 1067.94, N = 12 30804 43550 44561 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
OpenBenchmarking.org ms, Fewer Is Better PostgreSQL pgbench 14.0 Scaling Factor: 100 - Clients: 500 - Mode: Read Write - Average Latency Linux 5.16 Git Linux 5.14 Linux 5.15 4 8 12 16 20 SE +/- 2.66, N = 9 SE +/- 0.09, N = 9 SE +/- 0.26, N = 12 18.00 11.49 11.29 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm
ASTC Encoder ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.2 Preset: Thorough Linux 5.15 Linux 5.16 Git Linux 5.14 1.1351 2.2702 3.4053 4.5404 5.6755 SE +/- 0.0085, N = 3 SE +/- 0.0061, N = 3 SE +/- 0.0085, N = 3 5.0449 5.0332 5.0255 1. (CXX) g++ options: -O3 -flto -pthread
OpenBenchmarking.org Seconds, Fewer Is Better ASTC Encoder 3.2 Preset: Exhaustive Linux 5.15 Linux 5.16 Git Linux 5.14 3 6 9 12 15 SE +/- 0.0018, N = 3 SE +/- 0.0036, N = 3 SE +/- 0.0005, N = 3 9.3408 9.3375 9.3344 1. (CXX) g++ options: -O3 -flto -pthread
GIMP GIMP is an open-source image manipulaton program. This test profile will use the system-provided GIMP program otherwise on Windows relys upon a pre-packaged Windows binary from upstream GIMP.org. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.24 Test: resize Linux 5.14 Linux 5.15 Linux 5.16 Git 2 4 6 8 10 SE +/- 0.052, N = 3 SE +/- 0.074, N = 15 SE +/- 0.008, N = 3 8.735 8.697 7.545
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.24 Test: rotate Linux 5.15 Linux 5.14 Linux 5.16 Git 3 6 9 12 15 SE +/- 0.03, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 11.56 11.32 10.65
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.24 Test: auto-levels Linux 5.14 Linux 5.15 Linux 5.16 Git 4 8 12 16 20 SE +/- 0.17, N = 3 SE +/- 0.11, N = 3 SE +/- 0.20, N = 15 15.57 14.63 14.00
OpenBenchmarking.org Seconds, Fewer Is Better GIMP 2.10.24 Test: unsharp-mask Linux 5.14 Linux 5.16 Git Linux 5.15 5 10 15 20 25 SE +/- 0.40, N = 15 SE +/- 0.37, N = 15 SE +/- 0.23, N = 3 22.27 20.76 20.25
Hugin Hugin is an open-source, cross-platform panorama photo stitcher software package. This test profile times how long it takes to run the assistant and panorama photo stitching on a set of images. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Hugin Panorama Photo Assistant + Stitching Time Linux 5.14 Linux 5.16 Git Linux 5.15 11 22 33 44 55 SE +/- 0.26, N = 3 SE +/- 0.18, N = 3 SE +/- 0.52, N = 3 46.64 46.40 46.29
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: NUMA Linux 5.16 Git Linux 5.14 Linux 5.15 50 100 150 200 250 SE +/- 0.39, N = 3 SE +/- 0.34, N = 3 SE +/- 0.55, N = 3 199.23 235.28 243.97 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: MEMFD Linux 5.14 Linux 5.16 Git Linux 5.15 500 1000 1500 2000 2500 SE +/- 13.45, N = 3 SE +/- 2.76, N = 3 SE +/- 0.35, N = 3 2327.13 2414.38 2462.20 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Atomic Linux 5.16 Git Linux 5.14 Linux 5.15 40K 80K 120K 160K 200K SE +/- 125.74, N = 3 SE +/- 595.91, N = 3 SE +/- 354.03, N = 3 163786.83 164050.23 165643.13 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Crypto Linux 5.15 Linux 5.16 Git Linux 5.14 4K 8K 12K 16K 20K SE +/- 5.18, N = 3 SE +/- 1.62, N = 3 SE +/- 5.87, N = 3 17371.33 17371.42 17388.14 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Malloc Linux 5.15 Linux 5.14 400M 800M 1200M 1600M 2000M SE +/- 4580973.54, N = 3 SE +/- 2299751.81, N = 3 2020703794.62 2020739687.78 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
Test: RdRand
Linux 5.14: stress-ng: error: [3867659] No stress workers invoked (one or more were unsupported)
Linux 5.15: stress-ng: error: [2101491] No stress workers invoked (one or more were unsupported)
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Forking Linux 5.14 Linux 5.15 12K 24K 36K 48K 60K SE +/- 210.97, N = 3 SE +/- 368.30, N = 3 57083.55 57126.71 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: IO_uring Linux 5.14 Linux 5.15 200K 400K 600K 800K 1000K SE +/- 1086.76, N = 3 SE +/- 10688.68, N = 12 218218.47 840227.95 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: SENDFILE Linux 5.14 Linux 5.15 300K 600K 900K 1200K 1500K SE +/- 2091.66, N = 3 SE +/- 2228.52, N = 3 1300565.67 1307379.25 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: CPU Cache Linux 5.15 Linux 5.14 10 20 30 40 50 SE +/- 2.82, N = 12 SE +/- 2.66, N = 15 44.80 44.86 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: CPU Stress Linux 5.14 Linux 5.15 40K 80K 120K 160K 200K SE +/- 176.32, N = 3 SE +/- 631.34, N = 3 178339.70 180041.41 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Semaphores Linux 5.14 Linux 5.15 2M 4M 6M 8M 10M SE +/- 21401.85, N = 3 SE +/- 12321.83, N = 3 9832609.97 9850759.02 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Matrix Math Linux 5.15 Linux 5.14 60K 120K 180K 240K 300K SE +/- 425.68, N = 3 SE +/- 525.88, N = 3 262897.38 265236.91 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Vector Math Linux 5.14 Linux 5.15 80K 160K 240K 320K 400K SE +/- 46.03, N = 3 SE +/- 60.23, N = 3 383303.64 383450.16 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Memory Copying Linux 5.15 Linux 5.14 2K 4K 6K 8K 10K SE +/- 55.95, N = 3 SE +/- 22.01, N = 3 10338.86 10462.49 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Socket Activity Linux 5.14 Linux 5.15 10K 20K 30K 40K 50K SE +/- 188.94, N = 3 SE +/- 475.38, N = 3 44646.78 44932.67 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Context Switching Linux 5.15 Linux 5.14 6M 12M 18M 24M 30M SE +/- 305842.42, N = 5 SE +/- 199715.41, N = 3 27807770.52 29114551.20 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Glibc C String Functions Linux 5.15 Linux 5.14 2M 4M 6M 8M 10M SE +/- 52140.90, N = 3 SE +/- 10223.99, N = 3 9123960.66 9282586.48 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: Glibc Qsort Data Sorting Linux 5.14 Linux 5.15 200 400 600 800 1000 SE +/- 1.02, N = 3 SE +/- 0.39, N = 3 867.32 867.97 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.13.02 Test: System V Message Passing Linux 5.14 Linux 5.15 2M 4M 6M 8M 10M SE +/- 9347.98, N = 3 SE +/- 184885.66, N = 15 7223733.67 9104773.81 1. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -pthread -lc -latomic
Redis Memtier / Redis Benchmark Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. This test profile is for testing any system-installed Redis copy. The redis-server and redis-benchmark software must be present in PATH. To use a static / locally built version of Redis, run the pts/redis test profile instead. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better Redis Memtier / Redis Benchmark Test: LPUSH and LPOP: lpop Linux 5.16 Git Linux 5.14 Linux 5.15 300K 600K 900K 1200K 1500K SE +/- 2481.27, N = 15 SE +/- 6547.57, N = 3 SE +/- 26498.23, N = 12 300780.87 1224669.92 1234436.93 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre 2. Redis server v=6.0.15 sha=00000000:0 malloc=jemalloc-5.2.1 bits=64 build=4610f4c3acf7fb25
OpenBenchmarking.org Requests Per Second, More Is Better Redis Memtier / Redis Benchmark Test: LPUSH and LPOP: lpush Linux 5.16 Git Linux 5.14 Linux 5.15 300K 600K 900K 1200K 1500K SE +/- 2788.07, N = 3 SE +/- 2764.09, N = 3 SE +/- 17584.08, N = 3 313178.14 1282731.25 1322987.16 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre 2. Redis server v=6.0.15 sha=00000000:0 malloc=jemalloc-5.2.1 bits=64 build=4610f4c3acf7fb25
OpenBenchmarking.org Operations Per Second, More Is Better Redis Memtier / Redis Benchmark Test: GET Linux 5.16 Git Linux 5.15 Linux 5.14 200K 400K 600K 800K 1000K SE +/- 1110.87, N = 3 SE +/- 1840.53, N = 3 SE +/- 7100.00, N = 3 185507.15 390741.86 840518.83 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre 2. Redis server v=6.0.15 sha=00000000:0 malloc=jemalloc-5.2.1 bits=64 build=4610f4c3acf7fb25
OpenBenchmarking.org Operations Per Second, More Is Better Redis Memtier / Redis Benchmark Test: MIX Linux 5.16 Git Linux 5.15 Linux 5.14 110K 220K 330K 440K 550K SE +/- 2174.13, N = 3 SE +/- 5724.08, N = 12 SE +/- 9610.14, N = 12 172975.64 370509.97 497118.29 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre 2. Redis server v=6.0.15 sha=00000000:0 malloc=jemalloc-5.2.1 bits=64 build=4610f4c3acf7fb25
OpenBenchmarking.org Operations Per Second, More Is Better Redis Memtier / Redis Benchmark Test: SET Linux 5.16 Git Linux 5.14 Linux 5.15 70K 140K 210K 280K 350K SE +/- 2324.74, N = 3 SE +/- 3082.86, N = 12 SE +/- 3844.52, N = 12 183534.14 336736.34 349993.99 1. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre 2. Redis server v=6.0.15 sha=00000000:0 malloc=jemalloc-5.2.1 bits=64 build=4610f4c3acf7fb25
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: VGG19 - Device: CPU Linux 5.15 Linux 5.16 Git Linux 5.14 7 14 21 28 35 SE +/- 0.23, N = 3 SE +/- 0.36, N = 3 SE +/- 0.26, N = 3 27.19 27.27 30.27
FP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPU
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
FP16: No - Mode: Inference - Network: Mobilenet - Device: CPU
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
OpenBenchmarking.org FPS, More Is Better PlaidML FP16: No - Mode: Inference - Network: ResNet 50 - Device: CPU Linux 5.16 Git Linux 5.14 Linux 5.15 2 4 6 8 10 SE +/- 0.10, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 8.21 8.29 8.31
FP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPU
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
FP16: No - Mode: Inference - Network: Inception V3 - Device: CPU
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
FP16: No - Mode: Inference - Network: NASNer Large - Device: CPU
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
Blender Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles benchmark with various sample files. GPU computing via OpenCL, NVIDIA OptiX, and NVIDIA CUDA is supported. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: BMW27 - Compute: CPU-Only Linux 5.14 Linux 5.15 Linux 5.16 Git 6 12 18 24 30 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 SE +/- 0.08, N = 3 26.43 26.32 26.27
OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.92 Blend File: Barbershop - Compute: CPU-Only Linux 5.15 Linux 5.14 Linux 5.16 Git 20 40 60 80 100 SE +/- 0.13, N = 3 SE +/- 0.18, N = 3 SE +/- 0.13, N = 3 96.71 96.64 96.59
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times Linux 5.15 Linux 5.14 Linux 5.16 Git 200 400 600 800 1000 SE +/- 2.19, N = 3 SE +/- 2.33, N = 3 SE +/- 7.42, N = 3 882 878 867
nginx This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.21.1 Concurrent Requests: 1 Linux 5.16 Git Linux 5.15 Linux 5.14 13K 26K 39K 52K 65K SE +/- 182.67, N = 3 SE +/- 510.99, N = 8 SE +/- 517.34, N = 3 28680.17 58697.94 59479.82 1. (CC) gcc options: -lcrypt -lz -O3 -march=native
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.21.1 Concurrent Requests: 20 Linux 5.14 Linux 5.15 Linux 5.16 Git 30K 60K 90K 120K 150K SE +/- 2973.72, N = 15 SE +/- 3154.08, N = 15 SE +/- 3815.48, N = 15 113767.17 115953.22 118614.09 1. (CC) gcc options: -lcrypt -lz -O3 -march=native
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.21.1 Concurrent Requests: 100 Linux 5.16 Git Linux 5.14 Linux 5.15 20K 40K 60K 80K 100K SE +/- 1010.02, N = 3 SE +/- 386.34, N = 3 SE +/- 684.60, N = 3 90281.27 91191.83 94232.08 1. (CC) gcc options: -lcrypt -lz -O3 -march=native
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.21.1 Concurrent Requests: 200 Linux 5.14 Linux 5.15 Linux 5.16 Git 20K 40K 60K 80K 100K SE +/- 473.44, N = 3 SE +/- 121.73, N = 3 SE +/- 234.41, N = 3 92745.19 94210.56 96950.03 1. (CC) gcc options: -lcrypt -lz -O3 -march=native
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.21.1 Concurrent Requests: 500 Linux 5.14 Linux 5.15 Linux 5.16 Git 20K 40K 60K 80K 100K SE +/- 373.54, N = 3 SE +/- 797.11, N = 9 SE +/- 542.49, N = 3 95336.80 96204.68 101150.79 1. (CC) gcc options: -lcrypt -lz -O3 -march=native
OpenBenchmarking.org Requests Per Second, More Is Better nginx 1.21.1 Concurrent Requests: 1000 Linux 5.15 Linux 5.14 Linux 5.16 Git 20K 40K 60K 80K 100K SE +/- 57.77, N = 3 SE +/- 371.03, N = 3 SE +/- 759.19, N = 3 95536.75 96274.00 104669.51 1. (CC) gcc options: -lcrypt -lz -O3 -march=native
Nginx This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile relies upon nginx being present in the system PATH / already installed on the system if wanting to benchmark the operating system's provided nginx package. Use the pts/ngnx test profile if wanting to test a locally-built Nginx web server. Wrk is used for facilitating the HTTP web server load testing. Learn more via the OpenBenchmarking.org test page.
Test: Long Connection - Connections: 100
Linux 5.14: The test quit with a non-zero exit status.
Linux 5.15: The test quit with a non-zero exit status.
Linux 5.16 Git: The test quit with a non-zero exit status.
OpenBenchmarking.org Requests Per Second, More Is Better Nginx Test: Long Connection - Connections: 500 Linux 5.14 Linux 5.16 Git Linux 5.15 40K 80K 120K 160K 200K SE +/- 648.17, N = 3 SE +/- 868.35, N = 3 SE +/- 1530.94, N = 3 170923.03 173681.91 174938.99 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2 2. nginx version: nginx/1.18.0 (Ubuntu)
OpenBenchmarking.org Requests Per Second, More Is Better Nginx Test: Long Connection - Connections: 1000 Linux 5.16 Git Linux 5.14 Linux 5.15 40K 80K 120K 160K 200K SE +/- 460.60, N = 3 SE +/- 129.45, N = 3 SE +/- 289.67, N = 3 177123.95 178966.48 180288.02 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2 2. nginx version: nginx/1.18.0 (Ubuntu)
Test: Short Connection - Connections: 100
Linux 5.14: The test quit with a non-zero exit status.
Linux 5.15: The test quit with a non-zero exit status.
Linux 5.16 Git: The test quit with a non-zero exit status.
OpenBenchmarking.org Requests Per Second, More Is Better Nginx Test: Short Connection - Connections: 500 Linux 5.14 Linux 5.16 Git Linux 5.15 10K 20K 30K 40K 50K SE +/- 269.64, N = 15 SE +/- 266.17, N = 3 SE +/- 86.68, N = 3 42506.61 44473.37 44838.25 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2 2. nginx version: nginx/1.18.0 (Ubuntu)
OpenBenchmarking.org Requests Per Second, More Is Better Nginx Test: Short Connection - Connections: 1000 Linux 5.14 Linux 5.16 Git Linux 5.15 8K 16K 24K 32K 40K SE +/- 282.66, N = 11 SE +/- 401.75, N = 4 SE +/- 303.55, N = 3 37757.73 38639.24 39124.56 1. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2 2. nginx version: nginx/1.18.0 (Ubuntu)
ONNX Runtime ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.9.1 Model: yolov4 - Device: CPU Linux 5.16 Git Linux 5.14 Linux 5.15 60 120 180 240 300 SE +/- 6.75, N = 12 SE +/- 6.14, N = 12 SE +/- 9.15, N = 9 257 272 280 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.9.1 Model: fcn-resnet101-11 - Device: CPU Linux 5.15 Linux 5.16 Git Linux 5.14 40 80 120 160 200 SE +/- 2.02, N = 3 SE +/- 6.85, N = 12 SE +/- 9.26, N = 12 162 175 181 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.9.1 Model: shufflenet-v2-10 - Device: CPU Linux 5.14 Linux 5.15 Linux 5.16 Git 2K 4K 6K 8K 10K SE +/- 454.98, N = 12 SE +/- 437.90, N = 12 SE +/- 621.88, N = 12 6499 8026 8062 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
OpenBenchmarking.org Inferences Per Minute, More Is Better ONNX Runtime 1.9.1 Model: super-resolution-10 - Device: CPU Linux 5.15 Linux 5.16 Git Linux 5.14 1300 2600 3900 5200 6500 SE +/- 39.04, N = 3 SE +/- 477.70, N = 9 SE +/- 431.80, N = 12 5217 6002 6026 1. (CXX) g++ options: -ffunction-sections -fdata-sections -march=native -mtune=native -O3 -flto -fno-fat-lto-objects -ldl -lrt
Apache HTTP Server This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the Golang "Bombardier" program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.48 Concurrent Requests: 1 Linux 5.16 Git Linux 5.15 Linux 5.14 1200 2400 3600 4800 6000 SE +/- 74.71, N = 15 SE +/- 60.11, N = 15 SE +/- 51.91, N = 7 4020.66 5746.68 5787.27 1. (CC) gcc options: -shared -fPIC -O2
OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.48 Concurrent Requests: 20 Linux 5.16 Git Linux 5.15 Linux 5.14 8K 16K 24K 32K 40K SE +/- 478.84, N = 15 SE +/- 280.27, N = 3 SE +/- 873.54, N = 15 29625.79 31819.87 38522.89 1. (CC) gcc options: -shared -fPIC -O2
OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.48 Concurrent Requests: 100 Linux 5.14 Linux 5.16 Git Linux 5.15 20K 40K 60K 80K 100K SE +/- 117.37, N = 3 SE +/- 456.04, N = 3 SE +/- 218.37, N = 3 79423.25 84447.79 91230.65 1. (CC) gcc options: -shared -fPIC -O2
OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.48 Concurrent Requests: 200 Linux 5.16 Git Linux 5.14 Linux 5.15 30K 60K 90K 120K 150K SE +/- 628.63, N = 3 SE +/- 371.75, N = 3 SE +/- 1218.70, N = 3 97507.36 111041.99 117737.98 1. (CC) gcc options: -shared -fPIC -O2
OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.48 Concurrent Requests: 500 Linux 5.16 Git Linux 5.14 Linux 5.15 20K 40K 60K 80K 100K SE +/- 43.22, N = 3 SE +/- 305.75, N = 3 SE +/- 1002.07, N = 3 93482.06 98783.94 99259.70 1. (CC) gcc options: -shared -fPIC -O2
OpenBenchmarking.org Requests Per Second, More Is Better Apache HTTP Server 2.4.48 Concurrent Requests: 1000 Linux 5.16 Git Linux 5.15 Linux 5.14 20K 40K 60K 80K 100K SE +/- 727.25, N = 3 SE +/- 1436.71, N = 15 SE +/- 1649.90, N = 15 92727.08 106024.38 110183.33 1. (CC) gcc options: -shared -fPIC -O2
PyHPC Benchmarks PyHPC-Benchmarks is a suite of Python high performance computing benchmarks for execution on CPUs and GPUs using various popular Python HPC libraries. The PyHPC CPU-based benchmarks focus on sequential CPU performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: JAX - Project Size: 1048576 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0079 0.0158 0.0237 0.0316 0.0395 SE +/- 0.000, N = 15 SE +/- 0.000, N = 4 SE +/- 0.000, N = 9 0.035 0.035 0.035
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: JAX - Project Size: 1048576 - Benchmark: Isoneutral Mixing Linux 5.15 Linux 5.16 Git Linux 5.14 0.0394 0.0788 0.1182 0.1576 0.197 SE +/- 0.002, N = 15 SE +/- 0.002, N = 3 SE +/- 0.003, N = 15 0.175 0.174 0.169
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: JAX - Project Size: 4194304 - Benchmark: Equation of State Linux 5.16 Git Linux 5.14 Linux 5.15 0.0189 0.0378 0.0567 0.0756 0.0945 SE +/- 0.001, N = 3 SE +/- 0.001, N = 4 SE +/- 0.001, N = 3 0.084 0.083 0.082
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: JAX - Project Size: 4194304 - Benchmark: Isoneutral Mixing Linux 5.15 Linux 5.16 Git Linux 5.14 0.196 0.392 0.588 0.784 0.98 SE +/- 0.008, N = 3 SE +/- 0.012, N = 3 SE +/- 0.008, N = 13 0.871 0.869 0.825
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numba - Project Size: 1048576 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0126 0.0252 0.0378 0.0504 0.063 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.056 0.056 0.056
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numba - Project Size: 1048576 - Benchmark: Isoneutral Mixing Linux 5.14 Linux 5.15 Linux 5.16 Git 0.0673 0.1346 0.2019 0.2692 0.3365 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.299 0.297 0.295
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numba - Project Size: 4194304 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0504 0.1008 0.1512 0.2016 0.252 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.224 0.224 0.223
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numba - Project Size: 4194304 - Benchmark: Isoneutral Mixing Linux 5.15 Linux 5.16 Git Linux 5.14 0.2867 0.5734 0.8601 1.1468 1.4335 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 1.274 1.257 1.192
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0423 0.0846 0.1269 0.1692 0.2115 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 0.188 0.186 0.186
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 1048576 - Benchmark: Isoneutral Mixing Linux 5.16 Git Linux 5.15 Linux 5.14 0.1161 0.2322 0.3483 0.4644 0.5805 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.516 0.501 0.501
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Equation of State Linux 5.15 Linux 5.16 Git Linux 5.14 0.2419 0.4838 0.7257 0.9676 1.2095 SE +/- 0.000, N = 3 SE +/- 0.003, N = 3 SE +/- 0.001, N = 3 1.075 1.074 1.057
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Numpy - Project Size: 4194304 - Benchmark: Isoneutral Mixing Linux 5.15 Linux 5.16 Git Linux 5.14 0.4916 0.9832 1.4748 1.9664 2.458 SE +/- 0.002, N = 3 SE +/- 0.003, N = 3 SE +/- 0.003, N = 3 2.185 2.182 2.133
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Aesara - Project Size: 1048576 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0182 0.0364 0.0546 0.0728 0.091 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.081 0.081 0.081
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Aesara - Project Size: 1048576 - Benchmark: Isoneutral Mixing Linux 5.15 Linux 5.16 Git Linux 5.14 0.0736 0.1472 0.2208 0.2944 0.368 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.327 0.326 0.325
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Aesara - Project Size: 4194304 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0725 0.145 0.2175 0.29 0.3625 SE +/- 0.001, N = 3 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 0.322 0.322 0.321
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: Aesara - Project Size: 4194304 - Benchmark: Isoneutral Mixing Linux 5.15 Linux 5.16 Git Linux 5.14 0.3209 0.6418 0.9627 1.2836 1.6045 SE +/- 0.005, N = 3 SE +/- 0.002, N = 3 SE +/- 0.000, N = 3 1.426 1.421 1.417
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: PyTorch - Project Size: 1048576 - Benchmark: Equation of State Linux 5.15 Linux 5.14 Linux 5.16 Git 0.005 0.01 0.015 0.02 0.025 SE +/- 0.000, N = 15 SE +/- 0.000, N = 15 SE +/- 0.000, N = 13 0.022 0.022 0.021
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: PyTorch - Project Size: 1048576 - Benchmark: Isoneutral Mixing Linux 5.16 Git Linux 5.15 Linux 5.14 0.068 0.136 0.204 0.272 0.34 SE +/- 0.000, N = 3 SE +/- 0.001, N = 3 SE +/- 0.001, N = 3 0.302 0.302 0.298
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: PyTorch - Project Size: 4194304 - Benchmark: Equation of State Linux 5.15 Linux 5.16 Git Linux 5.14 0.0209 0.0418 0.0627 0.0836 0.1045 SE +/- 0.002, N = 12 SE +/- 0.001, N = 15 SE +/- 0.001, N = 15 0.093 0.091 0.090
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: PyTorch - Project Size: 4194304 - Benchmark: Isoneutral Mixing Linux 5.16 Git Linux 5.14 Linux 5.15 0.3186 0.6372 0.9558 1.2744 1.593 SE +/- 0.004, N = 3 SE +/- 0.013, N = 3 SE +/- 0.010, N = 3 1.416 1.413 1.404
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: TensorFlow - Project Size: 1048576 - Benchmark: Equation of State Linux 5.16 Git Linux 5.15 Linux 5.14 0.0065 0.013 0.0195 0.026 0.0325 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 SE +/- 0.000, N = 3 0.029 0.028 0.028
Device: CPU - Backend: TensorFlow - Project Size: 1048576 - Benchmark: Isoneutral Mixing
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
OpenBenchmarking.org Seconds, Fewer Is Better PyHPC Benchmarks 3.0 Device: CPU - Backend: TensorFlow - Project Size: 4194304 - Benchmark: Equation of State Linux 5.15 Linux 5.14 Linux 5.16 Git 0.0317 0.0634 0.0951 0.1268 0.1585 SE +/- 0.001, N = 3 SE +/- 0.002, N = 3 SE +/- 0.001, N = 3 0.141 0.140 0.138
Device: CPU - Backend: TensorFlow - Project Size: 4194304 - Benchmark: Isoneutral Mixing
Linux 5.14: Test failed to run.
Linux 5.15: Test failed to run.
Linux 5.16 Git: Test failed to run.
OpenCV This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.5.4 Test: Features 2D Linux 5.16 Git Linux 5.15 Linux 5.14 300K 600K 900K 1200K 1500K SE +/- 71578.33, N = 9 SE +/- 65802.90, N = 8 SE +/- 32498.69, N = 9 1278945 1251226 1226560 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.5.4 Test: Object Detection Linux 5.15 Linux 5.16 Git Linux 5.14 20K 40K 60K 80K 100K SE +/- 3198.15, N = 15 SE +/- 3448.20, N = 15 SE +/- 1222.05, N = 3 108769 101089 100090 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
OpenBenchmarking.org ms, Fewer Is Better OpenCV 4.5.4 Test: DNN - Deep Neural Network Linux 5.16 Git Linux 5.14 Linux 5.15 3K 6K 9K 12K 15K SE +/- 296.37, N = 15 SE +/- 388.98, N = 12 SE +/- 194.61, N = 15 12532 11572 11223 1. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared
Linux 5.14 Processor: 2 x AMD EPYC 75F3 32-Core @ 2.95GHz (64 Cores / 128 Threads), Motherboard: ASRockRack ROME2D16-2T (P3.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel 10G X550T
OS: Ubuntu 21.10, Kernel: 5.14.0-051400-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server, Vulkan: 1.1.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001114Python Notes: Python 3.9.7Security 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 IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 22 November 2021 13:23 by user phoronix.
Linux 5.15 Processor: 2 x AMD EPYC 75F3 32-Core @ 2.95GHz (64 Cores / 128 Threads), Motherboard: ASRockRack ROME2D16-2T (P3.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel 10G X550T
OS: Ubuntu 21.10, Kernel: 5.15.0-051500-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server, Vulkan: 1.1.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001114Python Notes: Python 3.9.7Security 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 IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 23 November 2021 13:45 by user phoronix.
Linux 5.16 Git Processor: 2 x AMD EPYC 75F3 32-Core @ 2.95GHz (64 Cores / 128 Threads), Motherboard: ASRockRack ROME2D16-2T (P3.10 BIOS), Chipset: AMD Starship/Matisse, Memory: 126GB, Disk: 1000GB Western Digital WD_BLACK SN850 1TB, Graphics: ASPEED, Audio: AMD Starship/Matisse, Monitor: VE228, Network: 2 x Intel 10G X550T
OS: Ubuntu 21.10, Kernel: 5.16.0-051600rc2daily20211124-generic (x86_64), Desktop: GNOME Shell 40.5, Display Server: X Server, Vulkan: 1.1.182, Compiler: GCC 11.2.0, File-System: ext4, Screen Resolution: 1920x1080
Kernel Notes: Transparent Huge Pages: madviseCompiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-ZPT0kp/gcc-11-11.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -vDisk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa001114Python Notes: Python 3.9.7Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected
Testing initiated at 24 November 2021 09:10 by user phoronix.