Ryzen 9 5900X LInux 5.11

AMD Ryzen 9 5900X 12-Core testing with a ASUS ROG CROSSHAIR VIII HERO (2702 BIOS) and Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB on Ubuntu 20.10 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2012231-HA-RYZEN959070
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

Audio Encoding 3 Tests
Bioinformatics 2 Tests
Web Browsers 1 Tests
Timed Code Compilation 4 Tests
C/C++ Compiler Tests 11 Tests
Compression Tests 2 Tests
CPU Massive 15 Tests
Creator Workloads 8 Tests
Database Test Suite 6 Tests
Desktop Graphics 5 Tests
Encoding 3 Tests
Fortran Tests 3 Tests
Game Development 2 Tests
Go Language Tests 2 Tests
HPC - High Performance Computing 11 Tests
Common Kernel Benchmarks 12 Tests
Machine Learning 4 Tests
Molecular Dynamics 3 Tests
MPI Benchmarks 4 Tests
Multi-Core 14 Tests
Networking Test Suite 2 Tests
NVIDIA GPU Compute 5 Tests
Intel oneAPI 4 Tests
OpenGL Demos Test Suite 3 Tests
OpenMPI Tests 4 Tests
Programmer / Developer System Benchmarks 9 Tests
Python 2 Tests
Scientific Computing 6 Tests
Server 8 Tests
Server CPU Tests 9 Tests
Single-Threaded 4 Tests
Unigine Test Suite 3 Tests
Vulkan Compute 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Linux 5.10.2
December 22 2020
  13 Hours, 58 Minutes
Linux 5.11 Git
December 23 2020
  15 Hours, 44 Minutes
Invert Hiding All Results Option
  14 Hours, 51 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

Linux 5.10.2

Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (2702 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211

OS: Ubuntu 20.10, Kernel: 5.10.2-051002-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
Graphics Notes: GLAMOR
Python Notes: Python 3.8.6
Security 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 5.11 Git

OS: Ubuntu 20.10, Kernel: 5.10.0-051000daily20201222-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Ryzen 9 5900X LInux 5.11OpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads)ASUS ROG CROSSHAIR VIII HERO (2702 BIOS)AMD Starship/Matisse16GB1000GB Sabrent Rocket 4.0 1TBSapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz)AMD Navi 10 HDMI AudioASUS VP28URealtek RTL8125 2.5GbE + Intel I211Ubuntu 20.105.10.2-051002-generic (x86_64)5.10.0-051000daily20201222-generic (x86_64)GNOME Shell 3.38.1X Server 1.20.9modesetting 1.20.94.6 Mesa 20.2.1 (LLVM 11.0.0)1.2.131GCC 10.2.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelsDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionRyzen 9 5900X LInux 5.11 BenchmarksSystem Logs- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 -v - NONE / errors=remount-ro,relatime,rw / Block Size: 4096- Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009- GLAMOR- Python 3.8.6- 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 5.10.2 vs. Linux 5.11 Git ComparisonPhoronix Test SuiteBaseline+17.4%+17.4%+34.8%+34.8%+52.2%+52.2%40.2%21.5%16.3%14.6%14.1%10.3%8.6%7.1%7.1%6%5.1%5%4%4%3.8%3.7%3.1%2.9%2.9%2.8%2.8%2.8%2.7%2.7%2.6%2.6%2.5%2.4%2.4%2.3%2.3%2.3%2.1%2.1%2.1%2.1%2%3840 x 2160 - Ultra69.4%U.U.D.S - 12860.1%Renderer2 - 3840 x 216053%3840 x 2160 - High43.8%Memcpy 1MB42.8%TCP - Connections/s - 163840 x 2160 - Ultimate37.5%100 - 1 - Read Only - Average Latency34.4%1 - 1 - Read Only - Average Latency28.6%100 - 1 - Read Only27.6%1 - 1 - Read Only24.6%HTTP - Bandwidth - 824.3%1920 x 108022.3%S.V.M.PRand Read20.1%Seq Read - Linux AIO - No - Yes - 2MB19.7%Seq Read - Linux AIO - No - Yes - 2MB19.6%M.M.B.S.T - f32 - CPU18.3%Hot Read18.2%2560 x 144017.8%Memorial17.3%Software CPU - 1920 x 108017.1%ARES-6 - Firefox17%Seek Rand16.6%NUMAHTTP - Bandwidth - 16SENDFILEB.B.B.S.4 - Software Only13.3%StyleBench - Firefox11.4%Jetstream - Firefox10.6%MotionMark - Firefox10.4%Create ProcessesM.P.P.B10%TCP - Bandwidth - 89.4%vklBenchmark8.6%1920 x 1080 - Fullscreen - OpenGL8.6%Epoll WaitPSPDFKit WASM - Firefox7.6%100 - 50 - Read Write - Average LatencySpeedometer - Firefox7.1%5201 - 30 Seconds - TCP - 107.1%100 - 50 - Read Write1920 x 1080 - Fullscreen - OpenGL7%64 - 10000 - 2,5000,1 - 100006.8%Jetstream 2 - Firefox6.4%B.B.B.S.1 - Software Only6.2%IP Shapes 3D - f32 - CPU6.1%TCP - Connections/s - 16%Unnamed Pipe - 204832 - Process5.4%16 - Process5.4%TCP - Latency - 165.3%1 - Process8 - Process5%1 - Thread16 - Thread4.9%4.5%IP Shapes 1D - f32 - CPU4.3%TCP Socket - 20484.3%8 - Thread4.2%Rand Read - Linux AIO - No - Yes - 4KBRand Read - Linux AIO - No - Yes - 4KB4 - 10000 - 2,5000,1 - 100004%1920 x 1080 - Fullscreen - Low - OpenGL3.9%OverwriteOverwriteC.B.S.A - f32 - CPU3.6%4 - Process3.6%Rand Write - Linux AIO - No - Yes - 4KB3.4%Rand Write - Linux AIO - No - Yes - 4KB3.3%2560 x 1440 - Fullscreen - OpenGL3.2%CPU Stress3.2%TCP - Connections/s - 85201 - 30 Seconds - TCP - 13.1%TCP - Latency - 13%Rand FillStatic OMP SpeedupSocket Activity2.8%M.L.U.sWAV To Ogg2.8%Rand Fill4 - Thread2.8%CPU - yolov4-tiny2.7%Rand Write - IO_uring - No - Yes - 4KBRand Write - IO_uring - No - Yes - 4KBCPU - regnety_400m2.6%Launch ProgramsUnnamed Pipe - 4096R.R.B2.5%2.5%Fill SyncContext Switching2.5%Time To CompileCPU - resnet182.4%Seq Fill2560 x 14402.4%U.U.D.S - 2048CPU CacheCanvasMark - Firefox2.3%CPU - googlenet2.3%Read While Writing2.3%Rand DeleteCPU - resnet502.2%Rand FillMemory AllocationsFIFO Named Pipe - 4096Create ThreadsOctane - Firefox2%3 - Compression SpeedXonoticIPC_benchmarkET: LegacyXonoticperf-benchEthrXonoticPostgreSQL pgbenchPostgreSQL pgbenchPostgreSQL pgbenchPostgreSQL pgbenchEthrTesseractStress-NGLevelDBFlexible IO TesterFlexible IO TesteroneDNNLevelDBTesseractIntel Open Image Denoiseyquake2SeleniumLevelDBStress-NGEthrStress-NGMPVSeleniumSeleniumSeleniumOSBenchHPC ChallengeEthrOpenVKLUnigine Valleyperf-benchSeleniumPostgreSQL pgbenchSeleniumiPerfPostgreSQL pgbenchUnigine HeavenInfluxDBSeleniumMPVoneDNNEthrIPC_benchmarkHackbenchHackbenchEthrHackbenchHackbenchHackbenchHackbenchVkFFToneDNNIPC_benchmarkHackbenchFlexible IO TesterFlexible IO TesterInfluxDBUnigine SuperpositionLevelDBLevelDBoneDNNHackbenchFlexible IO TesterFlexible IO TesterUnigine ValleyStress-NGEthriPerfEthrLevelDBCLOMPStress-NGBenchmarkMutexOgg Audio EncodingLevelDBNode.js V8 Web Tooling BenchmarkHackbenchNCNNFlexible IO TesterFlexible IO TesterNCNNOSBenchIPC_benchmarkHPC ChallengeKeyDBLevelDBStress-NGTimed Eigen CompilationNCNNFacebook RocksDBGLmark2IPC_benchmarkStress-NGSeleniumNCNNFacebook RocksDBLevelDBNCNNFacebook RocksDBOSBenchIPC_benchmarkOSBenchSeleniumLZ4 CompressionLinux 5.10.2Linux 5.11 Git

Ryzen 9 5900X LInux 5.11xonotic: 3840 x 2160 - Ultraipc-benchmark: Unnamed Unix Domain Socket - 128ethr: HTTP - Bandwidth - 8stress-ng: System V Message Passingonednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUtesseract: 2560 x 1440selenium: ARES-6 - Firefoxstress-ng: NUMAethr: HTTP - Bandwidth - 16stress-ng: SENDFILEmpv: Big Buck Bunny Sunflower 4K - Software Onlyselenium: StyleBench - Firefoxselenium: Jetstream - Firefoxhpcc: Max Ping Pong Bandwidthethr: TCP - Bandwidth - 8openvkl: vklBenchmarkunigine-valley: 1920 x 1080 - Fullscreen - OpenGLperf-bench: Epoll Waitselenium: PSPDFKit WASM - Firefoxpgbench: 100 - 50 - Read Write - Average Latencyselenium: Speedometer - Firefoxiperf: 5201 - 30 Seconds - TCP - 10pgbench: 100 - 50 - Read Writeunigine-heaven: 1920 x 1080 - Fullscreen - OpenGLinfluxdb: 64 - 10000 - 2,5000,1 - 10000selenium: Jetstream 2 - Firefoxmpv: Big Buck Bunny Sunflower 1080p - Software Onlyonednn: IP Shapes 3D - f32 - CPUethr: TCP - Connections/s - 1ipc-benchmark: Unnamed Pipe - 2048hackbench: 32 - Processhackbench: 16 - Processethr: TCP - Latency - 16hackbench: 1 - Processhackbench: 8 - Processhackbench: 1 - Threadhackbench: 16 - Threadvkfft: onednn: IP Shapes 1D - f32 - CPUipc-benchmark: TCP Socket - 2048hackbench: 8 - Threadfio: Rand Read - Linux AIO - No - Yes - 4KB - Default Test Directoryfio: Rand Read - Linux AIO - No - Yes - 4KB - Default Test Directoryinfluxdb: 4 - 10000 - 2,5000,1 - 10000unigine-super: 1920 x 1080 - Fullscreen - Low - OpenGLleveldb: Overwriteleveldb: Overwriteonednn: Convolution Batch Shapes Auto - f32 - CPUhackbench: 4 - Processfio: Rand Write - Linux AIO - No - Yes - 4KB - Default Test Directoryfio: Rand Write - Linux AIO - No - Yes - 4KB - Default Test Directoryunigine-valley: 2560 x 1440 - Fullscreen - OpenGLstress-ng: CPU Stressiperf: 5201 - 30 Seconds - TCP - 1leveldb: Rand Fillstress-ng: Socket Activitymutex: Mutex Lock Unlock spinlock_amdencode-ogg: WAV To Oggleveldb: Rand Fillnode-web-tooling: hackbench: 4 - Threadncnn: CPU - yolov4-tinyfio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directoryncnn: CPU - regnety_400mosbench: Launch Programsipc-benchmark: Unnamed Pipe - 4096hpcc: Rand Ring Bandwidthkeydb: leveldb: Fill Syncstress-ng: Context Switchingbuild-eigen: Time To Compilencnn: CPU - resnet18rocksdb: Seq Fillglmark2: 2560 x 1440ipc-benchmark: Unnamed Unix Domain Socket - 2048stress-ng: CPU Cacheselenium: CanvasMark - Firefoxncnn: CPU - googlenetrocksdb: Read While Writingleveldb: Rand Deletencnn: CPU - resnet50rocksdb: Rand Fillosbench: Memory Allocationsipc-benchmark: FIFO Named Pipe - 4096osbench: Create Threadsselenium: Octane - Firefoxcompress-lz4: 3 - Compression Speedleveldb: Seq Filliperf: 5201 - 30 Seconds - TCP - 5compress-lz4: 9 - Compression Speedipc-benchmark: FIFO Named Pipe - 2048leveldb: Seq Fillrocksdb: Rand Fill Syncfio: Seq Read - Linux AIO - No - Yes - 4KB - Default Test Directorymutex: Mutex Lock Unlock std::mutexfio: Seq Read - Linux AIO - No - Yes - 4KB - Default Test Directoryipc-benchmark: TCP Socket - 128ethr: TCP - Latency - 8ncnn: Vulkan GPU - resnet50ipc-benchmark: FIFO Named Pipe - 128ncnn: CPU - mnasnetstress-ng: Forkingncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - squeezenet_ssdmutex: Mutex Lock Unlock spinlockvkmark: 2560 x 1440xonotic: 3840 x 2160 - Lowonednn: Recurrent Neural Network Training - f32 - CPUhackbench: 2 - Threadncnn: Vulkan GPU - yolov4-tinyencode-wavpack: WAV To WavPackglmark2: 1920 x 1080perf-bench: Sched Pipemlpack: scikit_svmmutex: Mutex Lock Unlock pthread_mutexncnn: CPU-v2-v2 - mobilenet-v2ncnn: Vulkan GPU - regnety_400mstress-ng: Glibc C String Functionsembree: Pathtracer - Asian Dragonncnn: CPU - blazefacerocksdb: Rand Readwaifu2x-ncnn: 2x - 3 - Yesperf-bench: Syscall Basicncnn: Vulkan GPU - alexnetethr: UDP - Bandwidth - 16vkmark: 1920 x 1080hpcc: EP-DGEMMfio: Rand Read - Linux AIO - No - Yes - 2MB - Default Test Directoryfio: Rand Read - Linux AIO - No - Yes - 2MB - Default Test Directorybuild-linux-kernel: Time To Compileipc-benchmark: Unnamed Pipe - 128yquake2: Software CPU - 2560 x 1440hmmer: Pfam Database Searchethr: TCP - Bandwidth - 16ncnn: CPU - alexnetbuild-llvm: Time To Compilencnn: Vulkan GPU - mobilenetpgbench: 1 - 1 - Read Write - Average Latencyhpcc: Rand Ring Latencypgbench: 1 - 1 - Read Writencnn: CPU - vgg16mlpack: scikit_icapgbench: 100 - 50 - Read Only - Average Latencyfio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directoryncnn: Vulkan GPU - googlenetpgbench: 1 - 50 - Read Writevkmark: 3840 x 2160pgbench: 1 - 50 - Read Write - Average Latencyfio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directorymutex: Mutex Lock Unlock ticket_spinlockethr: HTTP - Bandwidth - 1hpcc: G-Rand Accessunigine-heaven: 2560 x 1440 - Fullscreen - OpenGLpgbench: 100 - 50 - Read Onlyhackbench: 2 - Processstress-ng: Semaphoresncnn: CPU - shufflenet-v2pgbench: 100 - 1 - Read Write - Average Latencyosbench: Create Filespgbench: 100 - 1 - Read Writestress-ng: Matrix Mathcompress-zstd: 3ncnn: CPU - efficientnet-b0sqlite-speedtest: Timed Time - Size 1,000ncnn: Vulkan GPU - blazefacencnn: Vulkan GPU - mnasnetembree: Pathtracer - Crownipc-benchmark: TCP Socket - 4096build2: Time To Compileencode-opus: WAV To Opus Encodehpcc: EP-STREAM Triadhpcc: G-HPLncnn: Vulkan GPU - shufflenet-v2selenium: WASM imageConvolute - Firefoxyquake2: Software CPU - 3840 x 2160mlpack: scikit_linearridgeregressionunigine-super: 1920 x 1080 - Fullscreen - Ultra - OpenGLembree: Pathtracer ISPC - Asian Dragonhimeno: Poisson Pressure Solvernamd: ATPase Simulation - 327,506 Atomsncnn: CPU - mobilenetstress-ng: Mallocperf-bench: Futex Lock-Pionednn: Deconvolution Batch shapes_3d - f32 - CPUncnn: Vulkan GPU - squeezenet_ssdmlpack: scikit_qdaselenium: Kraken - Firefoxyquake2: OpenGL 3.x - 1920 x 1080ncnn: Vulkan GPU-v3-v3 - mobilenet-v3embree: Pathtracer ISPC - Crownstress-ng: Memory Copyinglammps: Rhodopsin Proteinespeak: Text-To-Speech Synthesisstress-ng: MMAPhpcc: G-Ffteglmark2: 3840 x 2160ncnn: Vulkan GPU - efficientnet-b0ncnn: Vulkan GPU - vgg16realsr-ncnn: 4x - Nofio: Rand Write - IO_uring - No - Yes - 2MB - Default Test Directorystress-ng: MEMFDonednn: Recurrent Neural Network Inference - f32 - CPUyquake2: OpenGL 3.x - 2560 x 1440realsr-ncnn: 4x - Yesunigine-super: 1920 x 1080 - Fullscreen - High - OpenGLonednn: Deconvolution Batch shapes_1d - f32 - CPUstress-ng: Cryptopgbench: 1 - 50 - Read Onlycompress-lz4: 9 - Decompression Speedhpcg: fio: Seq Write - Linux AIO - No - Yes - 4KB - Default Test Directoryfio: Seq Write - Linux AIO - No - Yes - 2MB - Default Test Directoryfio: Rand Write - IO_uring - No - Yes - 2MB - Default Test Directorycompress-lz4: 1 - Decompression Speedfio: Seq Write - Linux AIO - No - Yes - 4KB - Default Test Directorytesseract: 3840 x 2160gromacs: Water Benchmarkncnn: Vulkan GPU - resnet18ethr: UDP - Bandwidth - 8fio: Seq Write - Linux AIO - No - Yes - 2MB - Default Test Directoryfio: Rand Write - Linux AIO - No - Yes - 2MB - Default Test Directoryselenium: WASM collisionDetection - Firefoxstress-ng: Vector Mathfio: Rand Read - IO_uring - No - Yes - 2MB - Default Test Directorycompress-lz4: 1 - Compression Speednumpy: perf-bench: Futex Hashfio: Seq Write - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Write - Linux AIO - No - Yes - 2MB - Default Test Directoryfio: Seq Write - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Read - IO_uring - No - Yes - 2MB - Default Test Directoryyquake2: OpenGL 3.x - 3840 x 2160perf-bench: Memset 1MBstress-ng: Atomicstress-ng: Glibc Qsort Data Sortingfio: Seq Read - IO_uring - No - Yes - 4KB - Default Test Directoryhpcc: G-Ptranscompress-lz4: 3 - Decompression Speedselenium: Maze Solver - Firefoxncnn: Vulkan GPU-v2-v2 - mobilenet-v2mutex: Semaphore Release And Acquiremutex: Mutex Lock Unlock std::mutexmutex: Shared Mutex Lock Sharedpgbench: 1 - 50 - Read Only - Average Latencycompress-zstd: 19simdjson: DistinctUserIDsimdjson: LargeRandiperf: 5201 - 30 Seconds - UDP - 10iperf: 5201 - 30 Seconds - UDP - 5iperf: 5201 - 30 Seconds - UDP - 1fio: Seq Read - IO_uring - No - Yes - 4KB - Default Test Directoryleveldb: Fill Syncselenium: MotionMark - Firefoxpgbench: 100 - 1 - Read Only - Average Latencypgbench: 100 - 1 - Read Onlypgbench: 1 - 1 - Read Only - Average Latencypgbench: 1 - 1 - Read Onlyoidn: Memorialclomp: Static OMP Speedupipc-benchmark: Unnamed Unix Domain Socket - 4096osbench: Create Processesperf-bench: Memcpy 1MBethr: TCP - Connections/s - 16ethr: TCP - Connections/s - 8ethr: TCP - Latency - 1yquake2: Software CPU - 1920 x 1080xonotic: 3840 x 2160 - Ultimatexonotic: 3840 x 2160 - Hightesseract: 1920 x 1080etlegacy: Renderer2 - 3840 x 2160fio: Seq Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Read - Linux AIO - No - Yes - 2MB - Default Test Directoryfio: Seq Read - Linux AIO - No - Yes - 2MB - Default Test Directoryfio: Seq Read - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Seq Read - IO_uring - No - Yes - 2MB - Default Test Directoryleveldb: Seek Randleveldb: Rand Readleveldb: Hot ReadLinux 5.10.2Linux 5.11 Git411.441125224016653969.5014501717.000.815800396.603350.62372.594906.65289659.24712.2992.7179.9532633.789147520.66666667236.14153.3226588833051.66212110279430095134.5371711157.685.3262077.197.620497805361938962.10930.43416.723.17315.3953.28031.111152043.61756259971515.79317294425001074910.6209.931.983.22715.78948.205554000216492.639810140.469277483.0609594.4621.914.67632.014.648.39619.89280271733316.6829.93345324367164.23009771374.861781.3186869871.4762.22612.9514081174495119714835.881601612.22358247077.70523.62128451252.89352023936458.8993712439869.6481.2269541867.06367406032.71282658066711.12268425780017.6123.8051833453.7881547.643.9613.9820.14573645.75598592958.334.73920.0711.175743817366816.359.134.2017.332076624.7720.72501.781063944836.4392251542710.45229572.00000000708119.203102380476852.8885289177105.381.024166055.1111111110.53414.52712.30.6210.43563161053.2439.350.128186112.241813227527.5774763338.671118.250.0460477.81603907514.5881813741.884.790.65110.152552153761027.315951.65.2842.8891.833.8119.7490185354191.0336.1062.1751244.886134.8324.149.32.5828.120.73385228.2768151.2903612.18205233129.846794.7882114.2059.44917.7980.44.0019.47931611.7311.04121.773374.666.0233520495.2953.4311.75040901106.361813.94976.880.05370.63.139663862.6352484013236.44.965305397782038204113564.72109300.69701.28513.05214200.2222222240834086296.4135272.19476712158.25508.5354394602038203940842380980.826.565906202668.16203.953280002.9656913150.24.34.228.1614.815.70.09545.20.880.5910.55.241.0512811.5399.960.032313160.0283655011.9117.2129043423.73457032.699619730005599917.47179.1329.7718612471.8253604488.2929278.47141672790156931451931386914.0959.9969.875242.926884414997863194.2117617104.630.965011336.610659.24433.315622.11330425.51628.8583.2162.7529671.383134820.31111111217.39141.1607156335561.5521139601632218125.7731601473.280.1671955.848.088137360383678565.44732.06917.603.02016.1653.12532.639145563.77483249308416.45717994603331033610.1202.133.180.26116.36298.501536000209589.74319828.419001880.7059330.5321.315.08932.915.058.63020.43287773646717.1229.17130825002634.12610752819.271738.5436705122.3460.76913.2614418024390122523636.721565312.50350335275.99424.15131189151.79667524439718.7189672391271.0179.6779362768.34374368333.31306157033311.32228418519117.3124.1851022243.8482828.504.0214.1920.44508636.64085032916.874.67320.3511.022734117141516.559.244.2517.132053352.3620.49381.801052501746.3712227794710.34227201.77777778700819.397202357472253.3925339526106.380.272164565.4222222210.44417.98412.200.6160.43915162353.6639.660.127184712.151826225927.3844730008.731125.790.0463577.30363933334.5591825250.874.820.64710.090916154660672.255917.85.3143.1251.843.7919.8462184480690.6066.1342.1850945.080674.8124.249.52.592820.80745210.8215131.2861212.22204569594.716774.8021514.1659.28920.1982.93.9919.52721615.6011.06621.822373.896.0354020455.2853.5311.77040841107.951811.34975.479.93970.53.135253867.4952421513220.74.971095403332040203913578.02111300.41441.28413.04214357.7777777840864083296.2135361.90477012151.51508.2554365822039203840862381980.426.573919202613.61204.003280672.9659913149.04.34.228.1614.815.70.09545.20.880.5910.55.241.0512811.5362.230.043245340.0362933110.1517.7128952421.52586022.9048061023415772418.00152.9239.8275114328.0886172399.1018182.07053332756131126291952391116.43312.00811.675OpenBenchmarking.org

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltraLinux 5.10.2Linux 5.11 Git90180270360450SE +/- 1.53, N = 3SE +/- 0.93, N = 3411.44242.93MIN: 181 / MAX: 618MIN: 152 / MAX: 555
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltraLinux 5.10.2Linux 5.11 Git70140210280350Min: 409.82 / Avg: 411.44 / Max: 414.49Min: 241.9 / Avg: 242.93 / Max: 244.79

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 128Linux 5.10.2Linux 5.11 Git500K1000K1500K2000K2500KSE +/- 23442.52, N = 3SE +/- 8883.31, N = 324016651499786
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 128Linux 5.10.2Linux 5.11 Git400K800K1200K1600K2000KMin: 2361124 / Avg: 2401665 / Max: 2442331Min: 1486217 / Avg: 1499786.33 / Max: 1516503

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 8Linux 5.10.2Linux 5.11 Git9001800270036004500SE +/- 11.76, N = 3SE +/- 6.23, N = 33969.503194.21MIN: 3740 / MAX: 4100MIN: 3130 / MAX: 3290
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 8Linux 5.10.2Linux 5.11 Git7001400210028003500Min: 3954.21 / Avg: 3969.5 / Max: 3992.63Min: 3183.16 / Avg: 3194.21 / Max: 3204.74

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: System V Message PassingLinux 5.11 GitLinux 5.10.24M8M12M16M20MSE +/- 196651.44, N = 3SE +/- 80454.79, N = 317617104.6314501717.001. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: System V Message PassingLinux 5.11 GitLinux 5.10.23M6M9M12M15MMin: 17395858.79 / Avg: 17617104.63 / Max: 18009335.56Min: 14395848.37 / Avg: 14501717 / Max: 14659593.181. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git0.21710.43420.65130.86841.0855SE +/- 0.002969, N = 3SE +/- 0.003270, N = 30.8158000.965011MIN: 0.79MIN: 0.931. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git246810Min: 0.81 / Avg: 0.82 / Max: 0.82Min: 0.96 / Avg: 0.97 / Max: 0.971. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git90180270360450SE +/- 4.77, N = 3SE +/- 1.97, N = 3396.60336.61
OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git70140210280350Min: 387.73 / Avg: 396.6 / Max: 404.05Min: 333.45 / Avg: 336.61 / Max: 340.23

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: ARES-6 - Browser: FirefoxLinux 5.10.2Linux 5.11 Git1326395265SE +/- 0.19, N = 3SE +/- 0.70, N = 350.6259.241. firefox 84.0
OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: ARES-6 - Browser: FirefoxLinux 5.10.2Linux 5.11 Git1224364860Min: 50.24 / Avg: 50.62 / Max: 50.81Min: 57.83 / Avg: 59.24 / Max: 601. firefox 84.0

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: NUMALinux 5.11 GitLinux 5.10.290180270360450SE +/- 1.31, N = 3SE +/- 1.60, N = 3433.31372.591. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: NUMALinux 5.11 GitLinux 5.10.280160240320400Min: 431.08 / Avg: 433.31 / Max: 435.61Min: 370.55 / Avg: 372.59 / Max: 375.751. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 16Linux 5.11 GitLinux 5.10.212002400360048006000SE +/- 4.72, N = 3SE +/- 6.59, N = 35622.114906.65MIN: 5440 / MAX: 5760MIN: 4680 / MAX: 5050
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 16Linux 5.11 GitLinux 5.10.210002000300040005000Min: 5613.68 / Avg: 5622.11 / Max: 5630Min: 4896.32 / Avg: 4906.65 / Max: 4918.89

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SENDFILELinux 5.11 GitLinux 5.10.270K140K210K280K350KSE +/- 274.26, N = 3SE +/- 318.35, N = 3330425.51289659.241. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SENDFILELinux 5.11 GitLinux 5.10.260K120K180K240K300KMin: 330065.27 / Avg: 330425.51 / Max: 330963.85Min: 289024.3 / Avg: 289659.24 / Max: 290017.641. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

MPV

MPV is an open-source, cross-platform media player. This test profile tests the frame-rate that can be achieved unsynchronized in a desynchronized mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterMPVVideo Input: Big Buck Bunny Sunflower 4K - Decode: Software OnlyLinux 5.10.2Linux 5.11 Git150300450600750SE +/- 4.75, N = 3SE +/- 2.05, N = 3712.29628.85MIN: 444.44 / MAX: 923.08MIN: 387.1 / MAX: 857.141. mpv 0.32.0
OpenBenchmarking.orgFPS, More Is BetterMPVVideo Input: Big Buck Bunny Sunflower 4K - Decode: Software OnlyLinux 5.10.2Linux 5.11 Git130260390520650Min: 703.13 / Avg: 712.29 / Max: 719.06Min: 624.76 / Avg: 628.85 / Max: 631.221. mpv 0.32.0

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns / Minute, More Is BetterSeleniumBenchmark: StyleBench - Browser: FirefoxLinux 5.10.2Linux 5.11 Git20406080100SE +/- 0.47, N = 3SE +/- 0.23, N = 392.783.21. firefox 84.0
OpenBenchmarking.orgRuns / Minute, More Is BetterSeleniumBenchmark: StyleBench - Browser: FirefoxLinux 5.10.2Linux 5.11 Git20406080100Min: 92 / Avg: 92.7 / Max: 93.6Min: 82.8 / Avg: 83.2 / Max: 83.61. firefox 84.0

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream - Browser: FirefoxLinux 5.10.2Linux 5.11 Git4080120160200SE +/- 0.60, N = 3SE +/- 1.39, N = 12179.95162.751. firefox 84.0
OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream - Browser: FirefoxLinux 5.10.2Linux 5.11 Git306090120150Min: 179.24 / Avg: 179.95 / Max: 181.15Min: 155.63 / Avg: 162.75 / Max: 170.881. firefox 84.0

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong BandwidthLinux 5.10.2Linux 5.11 Git7K14K21K28K35KSE +/- 107.50, N = 3SE +/- 161.16, N = 332633.7929671.381. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Max Ping Pong BandwidthLinux 5.10.2Linux 5.11 Git6K12K18K24K30KMin: 32459.1 / Avg: 32633.79 / Max: 32829.67Min: 29374.61 / Avg: 29671.38 / Max: 29928.71. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 8Linux 5.10.2Linux 5.11 Git30K60K90K120K150KSE +/- 1368.78, N = 3SE +/- 1861.95, N = 15147520.67134820.31MIN: 126280 / MAX: 158780MIN: 94320 / MAX: 163420
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 8Linux 5.10.2Linux 5.11 Git30K60K90K120K150KMin: 145199.33 / Avg: 147520.67 / Max: 149938Min: 121726 / Avg: 134820.31 / Max: 145125.33

OpenVKL

OpenVKL is the Intel Open Volume Kernel Library that offers high-performance volume computation kernels and part of the Intel oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkLinux 5.10.2Linux 5.11 Git50100150200250SE +/- 0.59, N = 3SE +/- 0.97, N = 3236.14217.39MIN: 1 / MAX: 931MIN: 1 / MAX: 929
OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkLinux 5.10.2Linux 5.11 Git4080120160200Min: 235 / Avg: 236.14 / Max: 237Min: 215.5 / Avg: 217.39 / Max: 218.75

Unigine Valley

This test calculates the average frame-rate within the Valley demo for the Unigine engine, released in February 2013. This engine is extremely demanding on the system's graphics card. Unigine Valley relies upon an OpenGL 3 core profile context. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Valley 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git306090120150SE +/- 0.31, N = 3SE +/- 1.01, N = 3153.32141.16
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Valley 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git306090120150Min: 152.98 / Avg: 153.32 / Max: 153.93Min: 139.16 / Avg: 141.16 / Max: 142.32

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitLinux 5.11 GitLinux 5.10.215K30K45K60K75KSE +/- 101.53, N = 3SE +/- 463.05, N = 371563658881. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitLinux 5.11 GitLinux 5.10.212K24K36K48K60KMin: 71457 / Avg: 71563 / Max: 71766Min: 64962 / Avg: 65888 / Max: 663631. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: FirefoxLinux 5.10.2Linux 5.11 Git8001600240032004000SE +/- 5.46, N = 3SE +/- 39.04, N = 4330535561. firefox 84.0
OpenBenchmarking.orgScore, Fewer Is BetterSeleniumBenchmark: PSPDFKit WASM - Browser: FirefoxLinux 5.10.2Linux 5.11 Git6001200180024003000Min: 3298 / Avg: 3305.33 / Max: 3316Min: 3493 / Avg: 3555.75 / Max: 36631. firefox 84.0

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.20.3740.7481.1221.4961.87SE +/- 0.003, N = 3SE +/- 0.002, N = 31.5521.6621. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.2246810Min: 1.55 / Avg: 1.55 / Max: 1.56Min: 1.66 / Avg: 1.66 / Max: 1.671. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRuns Per Minute, More Is BetterSeleniumBenchmark: Speedometer - Browser: FirefoxLinux 5.10.2Linux 5.11 Git306090120150SE +/- 1.47, N = 4SE +/- 0.88, N = 31211131. firefox 84.0
OpenBenchmarking.orgRuns Per Minute, More Is BetterSeleniumBenchmark: Speedometer - Browser: FirefoxLinux 5.10.2Linux 5.11 Git20406080100Min: 118 / Avg: 121 / Max: 124Min: 112 / Avg: 113.33 / Max: 1151. firefox 84.0

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 10Linux 5.10.2Linux 5.11 Git20K40K60K80K100KSE +/- 1106.82, N = 15SE +/- 691.66, N = 12102794960161. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 10Linux 5.10.2Linux 5.11 Git20K40K60K80K100KMin: 94238 / Avg: 102793.73 / Max: 107941Min: 94158 / Avg: 96015.83 / Max: 1025791. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read WriteLinux 5.11 GitLinux 5.10.27K14K21K28K35KSE +/- 52.46, N = 3SE +/- 34.41, N = 332218300951. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read WriteLinux 5.11 GitLinux 5.10.26K12K18K24K30KMin: 32159.74 / Avg: 32217.84 / Max: 32322.54Min: 30034.81 / Avg: 30094.5 / Max: 301541. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Unigine Heaven

This test calculates the average frame-rate within the Heaven demo for the Unigine engine. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 1920 x 1080 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git306090120150SE +/- 0.16, N = 3SE +/- 0.32, N = 3134.54125.77
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 1920 x 1080 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git306090120150Min: 134.22 / Avg: 134.54 / Max: 134.71Min: 125.17 / Avg: 125.77 / Max: 126.25

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.10.2Linux 5.11 Git400K800K1200K1600K2000KSE +/- 3810.04, N = 3SE +/- 21384.90, N = 31711157.61601473.2
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.10.2Linux 5.11 Git300K600K900K1200K1500KMin: 1703542.8 / Avg: 1711157.63 / Max: 1715210Min: 1560463.1 / Avg: 1601473.2 / Max: 1632493.5

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream 2 - Browser: FirefoxLinux 5.10.2Linux 5.11 Git20406080100SE +/- 0.15, N = 3SE +/- 0.54, N = 385.3380.171. firefox 84.0
OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: Jetstream 2 - Browser: FirefoxLinux 5.10.2Linux 5.11 Git1632486480Min: 85.09 / Avg: 85.33 / Max: 85.61Min: 79.14 / Avg: 80.17 / Max: 80.961. firefox 84.0

MPV

MPV is an open-source, cross-platform media player. This test profile tests the frame-rate that can be achieved unsynchronized in a desynchronized mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterMPVVideo Input: Big Buck Bunny Sunflower 1080p - Decode: Software OnlyLinux 5.10.2Linux 5.11 Git400800120016002000SE +/- 6.87, N = 3SE +/- 11.16, N = 32077.191955.84MIN: 1000 / MAX: 4000.16MIN: 923.08 / MAX: 4000.161. mpv 0.32.0
OpenBenchmarking.orgFPS, More Is BetterMPVVideo Input: Big Buck Bunny Sunflower 1080p - Decode: Software OnlyLinux 5.10.2Linux 5.11 Git400800120016002000Min: 2063.98 / Avg: 2077.19 / Max: 2087.08Min: 1933.6 / Avg: 1955.84 / Max: 1968.491. mpv 0.32.0

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git246810SE +/- 0.03242, N = 3SE +/- 0.01612, N = 37.620498.08813MIN: 7.32MIN: 7.951. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git3691215Min: 7.57 / Avg: 7.62 / Max: 7.68Min: 8.06 / Avg: 8.09 / Max: 8.111. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1Linux 5.10.2Linux 5.11 Git2K4K6K8K10KSE +/- 58.15, N = 15SE +/- 70.00, N = 378057360
OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 1Linux 5.10.2Linux 5.11 Git14002800420056007000Min: 7540 / Avg: 7805.33 / Max: 8520Min: 7250 / Avg: 7360 / Max: 7490

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 2048Linux 5.11 GitLinux 5.10.2800K1600K2400K3200K4000KSE +/- 32697.95, N = 15SE +/- 27847.34, N = 338367853619389
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 2048Linux 5.11 GitLinux 5.10.2700K1400K2100K2800K3500KMin: 3573442 / Avg: 3836784.6 / Max: 3971846Min: 3573356 / Avg: 3619389.33 / Max: 3669556

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessLinux 5.10.2Linux 5.11 Git1530456075SE +/- 0.05, N = 3SE +/- 0.15, N = 362.1165.451. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessLinux 5.10.2Linux 5.11 Git1326395265Min: 62.01 / Avg: 62.11 / Max: 62.16Min: 65.15 / Avg: 65.45 / Max: 65.651. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ProcessLinux 5.10.2Linux 5.11 Git714212835SE +/- 0.02, N = 3SE +/- 0.03, N = 330.4332.071. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ProcessLinux 5.10.2Linux 5.11 Git714212835Min: 30.4 / Avg: 30.43 / Max: 30.47Min: 32.03 / Avg: 32.07 / Max: 32.121. (CC) gcc options: -lpthread

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.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 16Linux 5.10.2Linux 5.11 Git48121620SE +/- 0.21, N = 4SE +/- 0.15, N = 1516.7217.60MIN: 6.95 / MAX: 21.64MIN: 6.12 / MAX: 24.22
OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 16Linux 5.10.2Linux 5.11 Git48121620Min: 16.14 / Avg: 16.72 / Max: 17.09Min: 16.37 / Avg: 17.6 / Max: 18.81

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ProcessLinux 5.11 GitLinux 5.10.20.71391.42782.14172.85563.5695SE +/- 0.003, N = 3SE +/- 0.014, N = 33.0203.1731. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ProcessLinux 5.11 GitLinux 5.10.2246810Min: 3.02 / Avg: 3.02 / Max: 3.03Min: 3.15 / Avg: 3.17 / Max: 3.21. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ProcessLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.02, N = 3SE +/- 0.00, N = 315.4016.171. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ProcessLinux 5.10.2Linux 5.11 Git48121620Min: 15.36 / Avg: 15.39 / Max: 15.42Min: 16.16 / Avg: 16.17 / Max: 16.171. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ThreadLinux 5.11 GitLinux 5.10.20.7381.4762.2142.9523.69SE +/- 0.010, N = 3SE +/- 0.006, N = 33.1253.2801. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ThreadLinux 5.11 GitLinux 5.10.2246810Min: 3.11 / Avg: 3.13 / Max: 3.14Min: 3.27 / Avg: 3.28 / Max: 3.291. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ThreadLinux 5.10.2Linux 5.11 Git816243240SE +/- 0.02, N = 3SE +/- 0.03, N = 331.1132.641. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ThreadLinux 5.10.2Linux 5.11 Git714212835Min: 31.09 / Avg: 31.11 / Max: 31.15Min: 32.58 / Avg: 32.64 / Max: 32.681. (CC) gcc options: -lpthread

VkFFT

VkFFT is a Fast Fourier Transform (FFT) Library that is GPU accelerated by means of the Vulkan API. The VkFFT benchmark runs FFT performance differences of many different sizes before returning an overall benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1Linux 5.10.2Linux 5.11 Git3K6K9K12K15KSE +/- 63.74, N = 3SE +/- 53.28, N = 315204145561. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgBenchmark Score, More Is BetterVkFFT 1.1.1Linux 5.10.2Linux 5.11 Git3K6K9K12K15KMin: 15082 / Avg: 15204 / Max: 15297Min: 14451 / Avg: 14556.33 / Max: 146231. (CXX) g++ options: -O3 -pthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git0.84931.69862.54793.39724.2465SE +/- 0.01821, N = 3SE +/- 0.04653, N = 33.617563.77483MIN: 3.36MIN: 3.511. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git246810Min: 3.58 / Avg: 3.62 / Max: 3.64Min: 3.69 / Avg: 3.77 / Max: 3.841. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 2048Linux 5.10.2Linux 5.11 Git600K1200K1800K2400K3000KSE +/- 24707.44, N = 3SE +/- 15786.00, N = 325997152493084
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 2048Linux 5.10.2Linux 5.11 Git500K1000K1500K2000K2500KMin: 2573116 / Avg: 2599714.67 / Max: 2649080Min: 2475053 / Avg: 2493084 / Max: 2524544

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ThreadLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.01, N = 3SE +/- 0.02, N = 315.7916.461. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ThreadLinux 5.10.2Linux 5.11 Git48121620Min: 15.78 / Avg: 15.79 / Max: 15.81Min: 16.43 / Avg: 16.46 / Max: 16.491. (CC) gcc options: -lpthread

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000SE +/- 3.18, N = 3SE +/- 28.55, N = 12179917291. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.230060090012001500Min: 1793 / Avg: 1799.33 / Max: 1803Min: 1449 / Avg: 1728.67 / Max: 18031. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2100K200K300K400K500KSE +/- 666.67, N = 3SE +/- 7295.18, N = 124603334425001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.280K160K240K320K400KMin: 459000 / Avg: 460333.33 / Max: 461000Min: 371000 / Avg: 442500 / Max: 4620001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.10.2Linux 5.11 Git200K400K600K800K1000KSE +/- 2020.53, N = 3SE +/- 451.41, N = 31074910.61033610.1
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.10.2Linux 5.11 Git200K400K600K800K1000KMin: 1071324.7 / Avg: 1074910.63 / Max: 1078317.1Min: 1032874.1 / Avg: 1033610.1 / Max: 1034430.9

Unigine Superposition

This test calculates the average frame-rate within the Superposition demo for the Unigine engine, released in 2017. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Low - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git50100150200250SE +/- 0.58, N = 3SE +/- 1.93, N = 3209.9202.1MAX: 314.8MAX: 317.9
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Low - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git4080120160200Min: 209.1 / Avg: 209.87 / Max: 211Min: 198.5 / Avg: 202.1 / Max: 205.1

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11 GitLinux 5.10.2816243240SE +/- 0.19, N = 3SE +/- 0.12, N = 333.131.91. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11 GitLinux 5.10.2714212835Min: 32.7 / Avg: 33.07 / Max: 33.3Min: 31.7 / Avg: 31.87 / Max: 32.11. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.40, N = 3SE +/- 0.30, N = 380.2683.231. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11 GitLinux 5.10.21632486480Min: 79.75 / Avg: 80.26 / Max: 81.06Min: 82.64 / Avg: 83.23 / Max: 83.621. (CXX) g++ options: -O3 -lsnappy -lpthread

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git48121620SE +/- 0.02, N = 3SE +/- 0.09, N = 315.7916.36MIN: 15.46MIN: 16.011. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git48121620Min: 15.75 / Avg: 15.79 / Max: 15.82Min: 16.22 / Avg: 16.36 / Max: 16.541. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ProcessLinux 5.10.2Linux 5.11 Git246810SE +/- 0.046, N = 3SE +/- 0.014, N = 38.2058.5011. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ProcessLinux 5.10.2Linux 5.11 Git3691215Min: 8.12 / Avg: 8.2 / Max: 8.28Min: 8.48 / Avg: 8.5 / Max: 8.531. (CC) gcc options: -lpthread

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git120K240K360K480K600KSE +/- 6806.86, N = 3SE +/- 4358.90, N = 35540005360001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git100K200K300K400K500KMin: 541000 / Avg: 554000 / Max: 564000Min: 529000 / Avg: 536000 / Max: 5440001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 26.42, N = 3SE +/- 16.90, N = 3216420951. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000Min: 2113 / Avg: 2163.67 / Max: 2202Min: 2068 / Avg: 2094.67 / Max: 21261. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

Unigine Valley

This test calculates the average frame-rate within the Valley demo for the Unigine engine, released in February 2013. This engine is extremely demanding on the system's graphics card. Unigine Valley relies upon an OpenGL 3 core profile context. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Valley 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git20406080100SE +/- 0.06, N = 3SE +/- 0.45, N = 392.6489.74
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Valley 1.0Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git20406080100Min: 92.55 / Avg: 92.64 / Max: 92.74Min: 88.92 / Avg: 89.74 / Max: 90.48

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU StressLinux 5.10.2Linux 5.11 Git2K4K6K8K10KSE +/- 37.01, N = 3SE +/- 114.61, N = 410140.469828.411. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU StressLinux 5.10.2Linux 5.11 Git2K4K6K8K10KMin: 10068.17 / Avg: 10140.46 / Max: 10190.41Min: 9547.41 / Avg: 9828.41 / Max: 10026.191. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 1Linux 5.10.2Linux 5.11 Git20K40K60K80K100KSE +/- 1250.48, N = 3SE +/- 1040.04, N = 1592774900181. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 1Linux 5.10.2Linux 5.11 Git16K32K48K64K80KMin: 91009 / Avg: 92774 / Max: 95191Min: 83886 / Avg: 90018.13 / Max: 958261. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.15, N = 3SE +/- 0.08, N = 380.7183.061. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11 GitLinux 5.10.21632486480Min: 80.5 / Avg: 80.71 / Max: 80.99Min: 82.93 / Avg: 83.06 / Max: 83.221. (CXX) g++ options: -O3 -lsnappy -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Socket ActivityLinux 5.10.2Linux 5.11 Git2K4K6K8K10KSE +/- 107.42, N = 3SE +/- 98.33, N = 159594.469330.531. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Socket ActivityLinux 5.10.2Linux 5.11 Git17003400510068008500Min: 9478.63 / Avg: 9594.46 / Max: 9809.08Min: 8794.94 / Avg: 9330.53 / Max: 10247.191. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlock_amdLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.03, N = 3SE +/- 0.00, N = 321.321.91. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlock_amdLinux 5.11 GitLinux 5.10.2510152025Min: 21.2 / Avg: 21.27 / Max: 21.3Min: 21.9 / Avg: 21.9 / Max: 21.91. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

Ogg Audio Encoding

This test times how long it takes to encode a sample WAV file to Ogg format using the reference Xiph.org tools/libraries. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOgg Audio Encoding 1.3.4WAV To OggLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.06, N = 3SE +/- 0.08, N = 314.6815.091. (CC) gcc options: -O2 -ffast-math -fsigned-char
OpenBenchmarking.orgSeconds, Fewer Is BetterOgg Audio Encoding 1.3.4WAV To OggLinux 5.10.2Linux 5.11 Git48121620Min: 14.56 / Avg: 14.68 / Max: 14.75Min: 15.01 / Avg: 15.09 / Max: 15.241. (CC) gcc options: -O2 -ffast-math -fsigned-char

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11 GitLinux 5.10.2816243240SE +/- 0.06, N = 3SE +/- 0.03, N = 332.932.01. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11 GitLinux 5.10.2714212835Min: 32.8 / Avg: 32.9 / Max: 33Min: 31.9 / Avg: 31.97 / Max: 321. (CXX) g++ options: -O3 -lsnappy -lpthread

Node.js V8 Web Tooling Benchmark

Running the V8 project's Web-Tooling-Benchmark under Node.js. The Web-Tooling-Benchmark stresses JavaScript-related workloads common to web developers like Babel and TypeScript and Babylon. This test profile can test the system's JavaScript performance with Node.js. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkLinux 5.11 GitLinux 5.10.248121620SE +/- 0.06, N = 3SE +/- 0.08, N = 315.0514.641. Nodejs v12.18.2
OpenBenchmarking.orgruns/s, More Is BetterNode.js V8 Web Tooling BenchmarkLinux 5.11 GitLinux 5.10.248121620Min: 14.96 / Avg: 15.05 / Max: 15.16Min: 14.48 / Avg: 14.64 / Max: 14.751. Nodejs v12.18.2

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ThreadLinux 5.10.2Linux 5.11 Git246810SE +/- 0.031, N = 3SE +/- 0.011, N = 38.3968.6301. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ThreadLinux 5.10.2Linux 5.11 Git3691215Min: 8.34 / Avg: 8.4 / Max: 8.43Min: 8.62 / Avg: 8.63 / Max: 8.651. (CC) gcc options: -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.10, N = 3SE +/- 0.48, N = 319.8920.43MIN: 19.59 / MAX: 20.33MIN: 19.79 / MAX: 21.711. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tinyLinux 5.10.2Linux 5.11 Git510152025Min: 19.7 / Avg: 19.89 / Max: 20.03Min: 19.93 / Avg: 20.43 / Max: 21.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.26001200180024003000SE +/- 24.12, N = 15SE +/- 32.59, N = 3287728021. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.25001000150020002500Min: 2668 / Avg: 2877 / Max: 2959Min: 2756 / Avg: 2802 / Max: 28651. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2160K320K480K640K800KSE +/- 6182.05, N = 15SE +/- 8089.77, N = 37364677173331. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2130K260K390K520K650KMin: 683000 / Avg: 736466.67 / Max: 757000Min: 706000 / Avg: 717333.33 / Max: 7330001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400mLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.36, N = 3SE +/- 0.07, N = 316.6817.12MIN: 15.85 / MAX: 18.66MIN: 16.91 / MAX: 19.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400mLinux 5.10.2Linux 5.11 Git48121620Min: 15.96 / Avg: 16.68 / Max: 17.12Min: 16.99 / Avg: 17.12 / Max: 17.231. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsLinux 5.11 GitLinux 5.10.2714212835SE +/- 0.23, N = 3SE +/- 0.33, N = 329.1729.931. (CC) gcc options: -lm
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsLinux 5.11 GitLinux 5.10.2714212835Min: 28.83 / Avg: 29.17 / Max: 29.62Min: 29.49 / Avg: 29.93 / Max: 30.581. (CC) gcc options: -lm

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 4096Linux 5.11 GitLinux 5.10.2500K1000K1500K2000K2500KSE +/- 24046.99, N = 3SE +/- 16121.23, N = 1525002632436716
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 4096Linux 5.11 GitLinux 5.10.2400K800K1200K1600K2000KMin: 2458558 / Avg: 2500263 / Max: 2541859Min: 2350371 / Avg: 2436716.13 / Max: 2581713

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring BandwidthLinux 5.10.2Linux 5.11 Git0.95181.90362.85543.80724.759SE +/- 0.01900, N = 3SE +/- 0.07138, N = 34.230094.126101. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: Random Ring BandwidthLinux 5.10.2Linux 5.11 Git246810Min: 4.2 / Avg: 4.23 / Max: 4.27Min: 4 / Avg: 4.13 / Max: 4.251. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

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.orgOps/sec, More Is BetterKeyDB 6.0.16Linux 5.10.2Linux 5.11 Git170K340K510K680K850KSE +/- 2699.75, N = 3SE +/- 1669.87, N = 3771374.86752819.271. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Linux 5.10.2Linux 5.11 Git130K260K390K520K650KMin: 766014.93 / Avg: 771374.86 / Max: 774619.85Min: 749806.26 / Avg: 752819.27 / Max: 755573.481. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.11 GitLinux 5.10.2400800120016002000SE +/- 3.72, N = 3SE +/- 1.19, N = 31738.541781.321. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.11 GitLinux 5.10.230060090012001500Min: 1732.47 / Avg: 1738.54 / Max: 1745.31Min: 1778.94 / Avg: 1781.32 / Max: 1782.531. (CXX) g++ options: -O3 -lsnappy -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingLinux 5.10.2Linux 5.11 Git1.5M3M4.5M6M7.5MSE +/- 92170.24, N = 3SE +/- 50290.98, N = 36869871.476705122.341. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingLinux 5.10.2Linux 5.11 Git1.2M2.4M3.6M4.8M6MMin: 6711792.22 / Avg: 6869871.47 / Max: 7031034.63Min: 6639756.82 / Avg: 6705122.34 / Max: 6804009.781. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

Timed Eigen Compilation

This test times how long it takes to build all Eigen examples. The Eigen examples are compiled serially. Eigen is a C++ template library for linear algebra. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To CompileLinux 5.11 GitLinux 5.10.21428425670SE +/- 0.05, N = 3SE +/- 0.33, N = 360.7762.23
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To CompileLinux 5.11 GitLinux 5.10.21224364860Min: 60.71 / Avg: 60.77 / Max: 60.86Min: 61.8 / Avg: 62.23 / Max: 62.87

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18Linux 5.10.2Linux 5.11 Git3691215SE +/- 0.04, N = 3SE +/- 0.19, N = 312.9513.26MIN: 12.82 / MAX: 13.21MIN: 12.97 / MAX: 22.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet18Linux 5.10.2Linux 5.11 Git48121620Min: 12.9 / Avg: 12.95 / Max: 13.04Min: 13.04 / Avg: 13.26 / Max: 13.641. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillLinux 5.11 GitLinux 5.10.2300K600K900K1200K1500KSE +/- 14455.88, N = 6SE +/- 5840.17, N = 3144180214081171. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillLinux 5.11 GitLinux 5.10.2200K400K600K800K1000KMin: 1403679 / Avg: 1441801.5 / Max: 1498256Min: 1400733 / Avg: 1408117.33 / Max: 14196471. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

GLmark2

This is a test of Linaro's glmark2 port, currently using the X11 OpenGL 2.0 target. GLmark2 is a basic OpenGL benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git1000200030004000500044954390

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 2048Linux 5.11 GitLinux 5.10.2300K600K900K1200K1500KSE +/- 2842.70, N = 3SE +/- 6366.08, N = 312252361197148
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 2048Linux 5.11 GitLinux 5.10.2200K400K600K800K1000KMin: 1222146 / Avg: 1225236 / Max: 1230914Min: 1184509 / Avg: 1197148.33 / Max: 1204797

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU CacheLinux 5.11 GitLinux 5.10.2816243240SE +/- 0.35, N = 15SE +/- 0.35, N = 1536.7235.881. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU CacheLinux 5.11 GitLinux 5.10.2816243240Min: 34.46 / Avg: 36.72 / Max: 38.86Min: 32.63 / Avg: 35.88 / Max: 38.031. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: CanvasMark - Browser: FirefoxLinux 5.10.2Linux 5.11 Git3K6K9K12K15KSE +/- 50.80, N = 3SE +/- 126.10, N = 316016156531. firefox 84.0
OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: CanvasMark - Browser: FirefoxLinux 5.10.2Linux 5.11 Git3K6K9K12K15KMin: 15946 / Avg: 16016.33 / Max: 16115Min: 15427 / Avg: 15652.67 / Max: 158631. firefox 84.0

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.04, N = 3SE +/- 0.30, N = 312.2212.50MIN: 12.07 / MAX: 13.83MIN: 12.07 / MAX: 20.491. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenetLinux 5.10.2Linux 5.11 Git48121620Min: 12.17 / Avg: 12.22 / Max: 12.29Min: 12.18 / Avg: 12.5 / Max: 13.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingLinux 5.10.2Linux 5.11 Git800K1600K2400K3200K4000KSE +/- 49577.80, N = 3SE +/- 10900.81, N = 3358247035033521. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingLinux 5.10.2Linux 5.11 Git600K1200K1800K2400K3000KMin: 3512469 / Avg: 3582470 / Max: 3678288Min: 3483244 / Avg: 3503352.33 / Max: 35207021. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.55, N = 3SE +/- 0.19, N = 375.9977.711. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteLinux 5.11 GitLinux 5.10.21530456075Min: 74.9 / Avg: 75.99 / Max: 76.56Min: 77.47 / Avg: 77.7 / Max: 78.081. (CXX) g++ options: -O3 -lsnappy -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50Linux 5.10.2Linux 5.11 Git612182430SE +/- 0.10, N = 3SE +/- 0.36, N = 323.6224.15MIN: 23.36 / MAX: 24.31MIN: 23.61 / MAX: 25.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50Linux 5.10.2Linux 5.11 Git612182430Min: 23.45 / Avg: 23.62 / Max: 23.78Min: 23.74 / Avg: 24.15 / Max: 24.871. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillLinux 5.11 GitLinux 5.10.2300K600K900K1200K1500KSE +/- 14903.56, N = 3SE +/- 16445.33, N = 3131189112845121. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillLinux 5.11 GitLinux 5.10.2200K400K600K800K1000KMin: 1289666 / Avg: 1311891.33 / Max: 1340205Min: 1260788 / Avg: 1284512.33 / Max: 13161031. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsLinux 5.11 GitLinux 5.10.21224364860SE +/- 0.04, N = 3SE +/- 0.59, N = 451.8052.891. (CC) gcc options: -lm
OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsLinux 5.11 GitLinux 5.10.21122334455Min: 51.76 / Avg: 51.8 / Max: 51.87Min: 51.62 / Avg: 52.89 / Max: 54.471. (CC) gcc options: -lm

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 4096Linux 5.11 GitLinux 5.10.2500K1000K1500K2000K2500KSE +/- 31508.66, N = 3SE +/- 24210.88, N = 1524439712393645
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 4096Linux 5.11 GitLinux 5.10.2400K800K1200K1600K2000KMin: 2384931 / Avg: 2443971.33 / Max: 2492572Min: 2283147 / Avg: 2393644.8 / Max: 2559057

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.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsLinux 5.11 GitLinux 5.10.2246810SE +/- 0.106917, N = 3SE +/- 0.064519, N = 158.7189678.8993711. (CC) gcc options: -lm
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsLinux 5.11 GitLinux 5.10.23691215Min: 8.58 / Avg: 8.72 / Max: 8.93Min: 8.36 / Avg: 8.9 / Max: 9.391. (CC) gcc options: -lm

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGeometric Mean, More Is BetterSeleniumBenchmark: Octane - Browser: FirefoxLinux 5.10.2Linux 5.11 Git5K10K15K20K25KSE +/- 109.36, N = 3SE +/- 151.27, N = 324398239121. firefox 84.0
OpenBenchmarking.orgGeometric Mean, More Is BetterSeleniumBenchmark: Octane - Browser: FirefoxLinux 5.10.2Linux 5.11 Git4K8K12K16K20KMin: 24184 / Avg: 24397.67 / Max: 24545Min: 23686 / Avg: 23911.67 / Max: 241991. firefox 84.0

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedLinux 5.11 GitLinux 5.10.21632486480SE +/- 0.86, N = 4SE +/- 0.76, N = 371.0169.641. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Compression SpeedLinux 5.11 GitLinux 5.10.21428425670Min: 68.69 / Avg: 71.01 / Max: 72.87Min: 68.68 / Avg: 69.64 / Max: 71.151. (CC) gcc options: -O3

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.54, N = 3SE +/- 0.21, N = 379.6881.231. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillLinux 5.11 GitLinux 5.10.21530456075Min: 79.02 / Avg: 79.68 / Max: 80.76Min: 80.83 / Avg: 81.23 / Max: 81.531. (CXX) g++ options: -O3 -lsnappy -lpthread

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 5Linux 5.10.2Linux 5.11 Git20K40K60K80K100KSE +/- 580.37, N = 3SE +/- 207.53, N = 395418936271. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: TCP - Parallel: 5Linux 5.10.2Linux 5.11 Git17K34K51K68K85KMin: 94594 / Avg: 95418 / Max: 96538Min: 93217 / Avg: 93627 / Max: 938881. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedLinux 5.11 GitLinux 5.10.21530456075SE +/- 0.68, N = 6SE +/- 0.22, N = 368.3467.061. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Compression SpeedLinux 5.11 GitLinux 5.10.21326395265Min: 67.11 / Avg: 68.34 / Max: 71.15Min: 66.72 / Avg: 67.06 / Max: 67.461. (CC) gcc options: -O3

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 2048Linux 5.11 GitLinux 5.10.2800K1600K2400K3200K4000KSE +/- 39537.52, N = 15SE +/- 32145.10, N = 837436833674060
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 2048Linux 5.11 GitLinux 5.10.2600K1200K1800K2400K3000KMin: 3432895 / Avg: 3743683.47 / Max: 3949241Min: 3560365 / Avg: 3674060.13 / Max: 3822987

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillLinux 5.11 GitLinux 5.10.2816243240SE +/- 0.22, N = 3SE +/- 0.07, N = 333.332.71. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillLinux 5.11 GitLinux 5.10.2714212835Min: 32.9 / Avg: 33.33 / Max: 33.6Min: 32.6 / Avg: 32.67 / Max: 32.81. (CXX) g++ options: -O3 -lsnappy -lpthread

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.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncLinux 5.11 GitLinux 5.10.23K6K9K12K15KSE +/- 10.35, N = 3SE +/- 9.81, N = 313061128261. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncLinux 5.11 GitLinux 5.10.22K4K6K8K10KMin: 13050 / Avg: 13061.33 / Max: 13082Min: 12809 / Avg: 12826 / Max: 128431. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git120K240K360K480K600KSE +/- 4112.68, N = 12SE +/- 5925.46, N = 35806675703331. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git100K200K300K400K500KMin: 564000 / Avg: 580666.67 / Max: 612000Min: 559000 / Avg: 570333.33 / Max: 5790001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.12, N = 3SE +/- 0.03, N = 311.111.31. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexLinux 5.10.2Linux 5.11 Git3691215Min: 10.9 / Avg: 11.1 / Max: 11.3Min: 11.3 / Avg: 11.33 / Max: 11.41. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 15.89, N = 12SE +/- 22.50, N = 3226822281. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000Min: 2204 / Avg: 2268.25 / Max: 2389Min: 2185 / Avg: 2228 / Max: 22611. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 128Linux 5.10.2Linux 5.11 Git900K1800K2700K3600K4500KSE +/- 24700.89, N = 3SE +/- 20170.75, N = 342578004185191
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 128Linux 5.10.2Linux 5.11 Git700K1400K2100K2800K3500KMin: 4208662 / Avg: 4257800.33 / Max: 4286782Min: 4144988 / Avg: 4185191.33 / Max: 4208182

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.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 8Linux 5.11 GitLinux 5.10.248121620SE +/- 0.18, N = 3SE +/- 0.19, N = 1517.3117.61MIN: 8.23 / MAX: 24.16MIN: 6.06 / MAX: 24.09
OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 8Linux 5.11 GitLinux 5.10.248121620Min: 16.95 / Avg: 17.31 / Max: 17.54Min: 16.61 / Avg: 17.61 / Max: 18.89

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet50Linux 5.10.2Linux 5.11 Git612182430SE +/- 0.05, N = 3SE +/- 0.33, N = 323.8024.18MIN: 23.54 / MAX: 31.91MIN: 23.59 / MAX: 25.191. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet50Linux 5.10.2Linux 5.11 Git612182430Min: 23.71 / Avg: 23.8 / Max: 23.86Min: 23.82 / Avg: 24.18 / Max: 24.841. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 128Linux 5.10.2Linux 5.11 Git1.1M2.2M3.3M4.4M5.5MSE +/- 61621.42, N = 4SE +/- 48203.31, N = 351833455102224
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 128Linux 5.10.2Linux 5.11 Git900K1800K2700K3600K4500KMin: 5050885 / Avg: 5183345 / Max: 5295087Min: 5014497 / Avg: 5102224.33 / Max: 5180709

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetLinux 5.10.2Linux 5.11 Git0.8641.7282.5923.4564.32SE +/- 0.02, N = 3SE +/- 0.04, N = 33.783.84MIN: 3.72 / MAX: 4.89MIN: 3.76 / MAX: 4.161. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnetLinux 5.10.2Linux 5.11 Git246810Min: 3.74 / Avg: 3.78 / Max: 3.81Min: 3.8 / Avg: 3.84 / Max: 3.911. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: ForkingLinux 5.11 GitLinux 5.10.220K40K60K80K100KSE +/- 365.70, N = 3SE +/- 158.65, N = 382828.5081547.641. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: ForkingLinux 5.11 GitLinux 5.10.214K28K42K56K70KMin: 82108.78 / Avg: 82828.5 / Max: 83301.05Min: 81346.42 / Avg: 81547.64 / Max: 81860.731. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3Linux 5.10.2Linux 5.11 Git0.90451.8092.71353.6184.5225SE +/- 0.02, N = 3SE +/- 0.03, N = 33.964.02MIN: 3.9 / MAX: 4.77MIN: 3.96 / MAX: 4.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3Linux 5.10.2Linux 5.11 Git246810Min: 3.92 / Avg: 3.96 / Max: 4Min: 3.98 / Avg: 4.02 / Max: 4.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.06, N = 3SE +/- 0.09, N = 313.9814.19MIN: 13.73 / MAX: 14.36MIN: 13.96 / MAX: 14.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssdLinux 5.10.2Linux 5.11 Git48121620Min: 13.87 / Avg: 13.98 / Max: 14.09Min: 14.1 / Avg: 14.19 / Max: 14.371. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlockLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.09, N = 3SE +/- 0.09, N = 320.120.41. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlockLinux 5.10.2Linux 5.11 Git510152025Min: 19.9 / Avg: 20.07 / Max: 20.2Min: 20.2 / Avg: 20.37 / Max: 20.51. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git10002000300040005000SE +/- 3.38, N = 3SE +/- 2.73, N = 3457345081. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git8001600240032004000Min: 4566 / Avg: 4572.67 / Max: 4577Min: 4503 / Avg: 4508.33 / Max: 45121. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: LowLinux 5.10.2Linux 5.11 Git140280420560700SE +/- 2.78, N = 3SE +/- 4.89, N = 3645.76636.64MIN: 390 / MAX: 1323MIN: 392 / MAX: 1290
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: LowLinux 5.10.2Linux 5.11 Git110220330440550Min: 642.82 / Avg: 645.76 / Max: 651.32Min: 628.87 / Avg: 636.64 / Max: 645.67

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.26001200180024003000SE +/- 37.58, N = 3SE +/- 24.19, N = 152916.872958.33MIN: 2849.92MIN: 2727.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.25001000150020002500Min: 2860.1 / Avg: 2916.87 / Max: 2987.91Min: 2733.94 / Avg: 2958.33 / Max: 3090.661. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ThreadLinux 5.11 GitLinux 5.10.21.06632.13263.19894.26525.3315SE +/- 0.005, N = 3SE +/- 0.009, N = 34.6734.7391. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ThreadLinux 5.11 GitLinux 5.10.2246810Min: 4.66 / Avg: 4.67 / Max: 4.68Min: 4.72 / Avg: 4.74 / Max: 4.751. (CC) gcc options: -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tinyLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.09, N = 3SE +/- 0.46, N = 320.0720.35MIN: 19.83 / MAX: 20.51MIN: 19.73 / MAX: 23.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: yolov4-tinyLinux 5.10.2Linux 5.11 Git510152025Min: 19.94 / Avg: 20.07 / Max: 20.25Min: 19.88 / Avg: 20.35 / Max: 21.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

WavPack Audio Encoding

This test times how long it takes to encode a sample WAV file to WavPack format with very high quality settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackLinux 5.11 GitLinux 5.10.23691215SE +/- 0.05, N = 5SE +/- 0.05, N = 511.0211.181. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackLinux 5.11 GitLinux 5.10.23691215Min: 10.86 / Avg: 11.02 / Max: 11.12Min: 11.03 / Avg: 11.18 / Max: 11.291. (CXX) g++ options: -rdynamic

GLmark2

This is a test of Linaro's glmark2 port, currently using the X11 OpenGL 2.0 target. GLmark2 is a basic OpenGL benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git1600320048006400800074387341

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeLinux 5.10.2Linux 5.11 Git40K80K120K160K200KSE +/- 2051.49, N = 3SE +/- 1147.85, N = 31736681714151. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeLinux 5.10.2Linux 5.11 Git30K60K90K120K150KMin: 171198 / Avg: 173667.67 / Max: 177740Min: 169632 / Avg: 171415.33 / Max: 1735591. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.14, N = 3SE +/- 0.04, N = 316.3516.55
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmLinux 5.10.2Linux 5.11 Git48121620Min: 16.15 / Avg: 16.35 / Max: 16.62Min: 16.5 / Avg: 16.55 / Max: 16.63

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock pthread_mutexLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.06, N = 3SE +/- 0.02, N = 39.139.241. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock pthread_mutexLinux 5.10.2Linux 5.11 Git3691215Min: 9 / Avg: 9.13 / Max: 9.19Min: 9.22 / Avg: 9.24 / Max: 9.271. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2Linux 5.10.2Linux 5.11 Git0.95631.91262.86893.82524.7815SE +/- 0.03, N = 3SE +/- 0.03, N = 34.204.25MIN: 4.1 / MAX: 4.92MIN: 4.16 / MAX: 6.051. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2Linux 5.10.2Linux 5.11 Git246810Min: 4.15 / Avg: 4.2 / Max: 4.23Min: 4.22 / Avg: 4.25 / Max: 4.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400mLinux 5.11 GitLinux 5.10.248121620SE +/- 0.05, N = 3SE +/- 0.12, N = 317.1317.33MIN: 16.96 / MAX: 17.68MIN: 17.02 / MAX: 18.751. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: regnety_400mLinux 5.11 GitLinux 5.10.248121620Min: 17.03 / Avg: 17.13 / Max: 17.21Min: 17.1 / Avg: 17.33 / Max: 17.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc C String FunctionsLinux 5.10.2Linux 5.11 Git400K800K1200K1600K2000KSE +/- 12611.55, N = 3SE +/- 18853.69, N = 32076624.772053352.361. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc C String FunctionsLinux 5.10.2Linux 5.11 Git400K800K1200K1600K2000KMin: 2052746.35 / Avg: 2076624.77 / Max: 2095600.97Min: 2023455.25 / Avg: 2053352.36 / Max: 2088200.851. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian DragonLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.09, N = 3SE +/- 0.10, N = 320.7320.49MIN: 20.5 / MAX: 21.06MIN: 20.22 / MAX: 20.9
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian DragonLinux 5.10.2Linux 5.11 Git510152025Min: 20.58 / Avg: 20.73 / Max: 20.88Min: 20.33 / Avg: 20.49 / Max: 20.68

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceLinux 5.10.2Linux 5.11 Git0.4050.811.2151.622.025SE +/- 0.02, N = 3SE +/- 0.01, N = 31.781.80MIN: 1.73 / MAX: 1.92MIN: 1.78 / MAX: 2.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazefaceLinux 5.10.2Linux 5.11 Git246810Min: 1.75 / Avg: 1.78 / Max: 1.8Min: 1.79 / Avg: 1.8 / Max: 1.821. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadLinux 5.10.2Linux 5.11 Git20M40M60M80M100MSE +/- 126364.22, N = 3SE +/- 72529.36, N = 31063944831052501741. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadLinux 5.10.2Linux 5.11 Git20M40M60M80M100MMin: 106262798 / Avg: 106394483.33 / Max: 106647136Min: 105113690 / Avg: 105250174.33 / Max: 1053609671. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Waifu2x-NCNN Vulkan

Waifu2x-NCNN is an NCNN neural network implementation of the Waifu2x converter project and accelerated using the Vulkan API. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesLinux 5.11 GitLinux 5.10.2246810SE +/- 0.012, N = 3SE +/- 0.009, N = 36.3716.439
OpenBenchmarking.orgSeconds, Fewer Is BetterWaifu2x-NCNN Vulkan 20200818Scale: 2x - Denoise: 3 - TAA: YesLinux 5.11 GitLinux 5.10.23691215Min: 6.35 / Avg: 6.37 / Max: 6.38Min: 6.42 / Avg: 6.44 / Max: 6.45

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicLinux 5.10.2Linux 5.11 Git5M10M15M20M25MSE +/- 103801.56, N = 3SE +/- 5682.55, N = 322515427222779471. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicLinux 5.10.2Linux 5.11 Git4M8M12M16M20MMin: 22350294 / Avg: 22515427 / Max: 22706955Min: 22266879 / Avg: 22277947.33 / Max: 222857161. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnetLinux 5.11 GitLinux 5.10.23691215SE +/- 0.01, N = 3SE +/- 0.11, N = 310.3410.45MIN: 10.26 / MAX: 10.55MIN: 10.28 / MAX: 12.351. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: alexnetLinux 5.11 GitLinux 5.10.23691215Min: 10.32 / Avg: 10.34 / Max: 10.35Min: 10.34 / Avg: 10.45 / Max: 10.671. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 16Linux 5.10.2Linux 5.11 Git50K100K150K200K250KSE +/- 455.90, N = 3SE +/- 226.58, N = 3229572.00227201.78MIN: 226350 / MAX: 233820MIN: 224930 / MAX: 234160
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 16Linux 5.10.2Linux 5.11 Git40K80K120K160K200KMin: 228955.33 / Avg: 229572 / Max: 230462Min: 226832.67 / Avg: 227201.78 / Max: 227614

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git15003000450060007500SE +/- 3.06, N = 3SE +/- 2.96, N = 3708170081. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git12002400360048006000Min: 7075 / Avg: 7081 / Max: 7085Min: 7004 / Avg: 7008.33 / Max: 70141. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMMLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.10, N = 3SE +/- 0.13, N = 319.4019.201. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: EP-DGEMMLinux 5.11 GitLinux 5.10.2510152025Min: 19.24 / Avg: 19.4 / Max: 19.59Min: 18.99 / Avg: 19.2 / Max: 19.441. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 0.33, N = 3SE +/- 14.84, N = 3238023571. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000Min: 2380 / Avg: 2380.33 / Max: 2381Min: 2330 / Avg: 2357.33 / Max: 23811. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git10002000300040005000SE +/- 0.88, N = 3SE +/- 29.94, N = 3476847221. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git8001600240032004000Min: 4767 / Avg: 4768.33 / Max: 4770Min: 4667 / Avg: 4722 / Max: 47701. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileLinux 5.10.2Linux 5.11 Git1224364860SE +/- 0.33, N = 3SE +/- 0.37, N = 352.8953.39
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileLinux 5.10.2Linux 5.11 Git1122334455Min: 52.45 / Avg: 52.89 / Max: 53.53Min: 52.95 / Avg: 53.39 / Max: 54.12

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 128Linux 5.11 GitLinux 5.10.21.1M2.2M3.3M4.4M5.5MSE +/- 57463.04, N = 15SE +/- 46297.15, N = 1553395265289177
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 128Linux 5.11 GitLinux 5.10.2900K1800K2700K3600K4500KMin: 5012417 / Avg: 5339526.33 / Max: 5831248Min: 5079715 / Avg: 5289177.2 / Max: 5591361

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 2560 x 1440Linux 5.11 GitLinux 5.10.220406080100SE +/- 1.46, N = 3SE +/- 0.60, N = 3106.3105.31. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 2560 x 1440Linux 5.11 GitLinux 5.10.220406080100Min: 104.2 / Avg: 106.3 / Max: 109.1Min: 104.1 / Avg: 105.27 / Max: 106.11. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.16, N = 3SE +/- 0.14, N = 380.2781.021. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchLinux 5.11 GitLinux 5.10.21530456075Min: 79.95 / Avg: 80.27 / Max: 80.45Min: 80.88 / Avg: 81.02 / Max: 81.311. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 16Linux 5.10.2Linux 5.11 Git40K80K120K160K200KSE +/- 2208.72, N = 3SE +/- 1344.75, N = 15166055.11164565.42MIN: 160160 / MAX: 174000MIN: 151060 / MAX: 174580
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 16Linux 5.10.2Linux 5.11 Git30K60K90K120K150KMin: 163557.33 / Avg: 166055.11 / Max: 170459.33Min: 156081.33 / Avg: 164565.42 / Max: 171794

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetLinux 5.11 GitLinux 5.10.23691215SE +/- 0.11, N = 3SE +/- 0.08, N = 310.4410.53MIN: 10.28 / MAX: 10.71MIN: 10.3 / MAX: 18.311. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnetLinux 5.11 GitLinux 5.10.23691215Min: 10.33 / Avg: 10.44 / Max: 10.66Min: 10.37 / Avg: 10.53 / Max: 10.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileLinux 5.10.2Linux 5.11 Git90180270360450SE +/- 3.75, N = 3SE +/- 4.03, N = 3414.53417.98
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileLinux 5.10.2Linux 5.11 Git70140210280350Min: 410.49 / Avg: 414.53 / Max: 422.02Min: 410.77 / Avg: 417.98 / Max: 424.71

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenetLinux 5.11 GitLinux 5.10.23691215SE +/- 0.07, N = 3SE +/- 0.03, N = 312.2012.30MIN: 12.03 / MAX: 14.49MIN: 12.12 / MAX: 12.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mobilenetLinux 5.11 GitLinux 5.10.248121620Min: 12.13 / Avg: 12.2 / Max: 12.34Min: 12.2 / Avg: 12.27 / Max: 12.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.20.13970.27940.41910.55880.6985SE +/- 0.001, N = 3SE +/- 0.001, N = 30.6160.6211. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.2246810Min: 0.62 / Avg: 0.62 / Max: 0.62Min: 0.62 / Avg: 0.62 / Max: 0.621. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring LatencyLinux 5.10.2Linux 5.11 Git0.09880.19760.29640.39520.494SE +/- 0.00200, N = 3SE +/- 0.00254, N = 30.435630.439151. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.5.0Test / Class: Random Ring LatencyLinux 5.10.2Linux 5.11 Git12345Min: 0.43 / Avg: 0.44 / Max: 0.44Min: 0.44 / Avg: 0.44 / Max: 0.441. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read WriteLinux 5.11 GitLinux 5.10.230060090012001500SE +/- 1.80, N = 3SE +/- 3.56, N = 3162316101. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read WriteLinux 5.11 GitLinux 5.10.230060090012001500Min: 1620.04 / Avg: 1622.5 / Max: 1626.01Min: 1602.8 / Avg: 1609.57 / Max: 1614.861. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16Linux 5.10.2Linux 5.11 Git1224364860SE +/- 0.20, N = 3SE +/- 0.15, N = 353.2453.66MIN: 52.36 / MAX: 60.28MIN: 52.97 / MAX: 54.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg16Linux 5.10.2Linux 5.11 Git1122334455Min: 52.97 / Avg: 53.24 / Max: 53.62Min: 53.45 / Avg: 53.66 / Max: 53.951. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaLinux 5.10.2Linux 5.11 Git918273645SE +/- 0.08, N = 3SE +/- 0.16, N = 339.3539.66
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaLinux 5.10.2Linux 5.11 Git816243240Min: 39.24 / Avg: 39.35 / Max: 39.51Min: 39.36 / Avg: 39.66 / Max: 39.9

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average LatencyLinux 5.11 GitLinux 5.10.20.02880.05760.08640.11520.144SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1270.1281. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average LatencyLinux 5.11 GitLinux 5.10.212345Min: 0.13 / Avg: 0.13 / Max: 0.13Min: 0.13 / Avg: 0.13 / Max: 0.131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000SE +/- 2.40, N = 3SE +/- 5.90, N = 3186118471. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git30060090012001500Min: 1856 / Avg: 1860.67 / Max: 1864Min: 1840 / Avg: 1847.33 / Max: 18591. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenetLinux 5.11 GitLinux 5.10.23691215SE +/- 0.00, N = 3SE +/- 0.07, N = 312.1512.24MIN: 12.04 / MAX: 12.55MIN: 12.06 / MAX: 12.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: googlenetLinux 5.11 GitLinux 5.10.248121620Min: 12.15 / Avg: 12.15 / Max: 12.16Min: 12.15 / Avg: 12.24 / Max: 12.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read WriteLinux 5.11 GitLinux 5.10.2400800120016002000SE +/- 1.95, N = 3SE +/- 0.90, N = 3182618131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read WriteLinux 5.11 GitLinux 5.10.230060090012001500Min: 1824.09 / Avg: 1826.18 / Max: 1830.07Min: 1811.74 / Avg: 1813.43 / Max: 1814.821. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

VKMark

VKMark is a collection of Vulkan tests/benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git5001000150020002500SE +/- 0.67, N = 3SE +/- 3.51, N = 3227522591. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF
OpenBenchmarking.orgVKMark Score, More Is BetterVKMark 2020-05-21Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git400800120016002000Min: 2274 / Avg: 2274.67 / Max: 2276Min: 2255 / Avg: 2259 / Max: 22661. (CXX) g++ options: -pthread -ldl -pipe -std=c++14 -MD -MQ -MF

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.2612182430SE +/- 0.03, N = 3SE +/- 0.01, N = 327.3827.581. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.2612182430Min: 27.33 / Avg: 27.38 / Max: 27.42Min: 27.56 / Avg: 27.58 / Max: 27.61. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git100K200K300K400K500KSE +/- 666.67, N = 3SE +/- 1527.53, N = 34763334730001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git80K160K240K320K400KMin: 475000 / Avg: 476333.33 / Max: 477000Min: 471000 / Avg: 473000 / Max: 4760001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock ticket_spinlockLinux 5.10.2Linux 5.11 Git246810SE +/- 0.08, N = 3SE +/- 0.02, N = 38.678.731. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock ticket_spinlockLinux 5.10.2Linux 5.11 Git3691215Min: 8.52 / Avg: 8.67 / Max: 8.79Min: 8.7 / Avg: 8.73 / Max: 8.751. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1Linux 5.11 GitLinux 5.10.22004006008001000SE +/- 1.39, N = 3SE +/- 0.35, N = 31125.791118.25MIN: 1100 / MAX: 1150MIN: 1100 / MAX: 1140
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1Linux 5.11 GitLinux 5.10.22004006008001000Min: 1123.68 / Avg: 1125.79 / Max: 1128.42Min: 1117.89 / Avg: 1118.25 / Max: 1118.95

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random AccessLinux 5.11 GitLinux 5.10.20.01040.02080.03120.04160.052SE +/- 0.00016, N = 3SE +/- 0.00030, N = 30.046350.046041. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-Random AccessLinux 5.11 GitLinux 5.10.212345Min: 0.05 / Avg: 0.05 / Max: 0.05Min: 0.05 / Avg: 0.05 / Max: 0.051. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

Unigine Heaven

This test calculates the average frame-rate within the Heaven demo for the Unigine engine. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git20406080100SE +/- 0.49, N = 3SE +/- 0.45, N = 377.8277.30
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Heaven 4.0Resolution: 2560 x 1440 - Mode: Fullscreen - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git1530456075Min: 76.83 / Avg: 77.82 / Max: 78.36Min: 76.4 / Avg: 77.3 / Max: 77.8

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyLinux 5.11 GitLinux 5.10.280K160K240K320K400KSE +/- 411.53, N = 3SE +/- 765.65, N = 33933333907511. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyLinux 5.11 GitLinux 5.10.270K140K210K280K350KMin: 392522.38 / Avg: 393333.44 / Max: 393860.2Min: 389274.33 / Avg: 390750.54 / Max: 391841.131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ProcessLinux 5.11 GitLinux 5.10.21.03232.06463.09694.12925.1615SE +/- 0.005, N = 3SE +/- 0.001, N = 34.5594.5881. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ProcessLinux 5.11 GitLinux 5.10.2246810Min: 4.55 / Avg: 4.56 / Max: 4.57Min: 4.59 / Avg: 4.59 / Max: 4.591. (CC) gcc options: -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SemaphoresLinux 5.11 GitLinux 5.10.2400K800K1200K1600K2000KSE +/- 4444.66, N = 3SE +/- 8971.20, N = 31825250.871813741.881. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SemaphoresLinux 5.11 GitLinux 5.10.2300K600K900K1200K1500KMin: 1817559.85 / Avg: 1825250.87 / Max: 1832956.6Min: 1800311.27 / Avg: 1813741.88 / Max: 1830760.661. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2Linux 5.10.2Linux 5.11 Git1.08452.1693.25354.3385.4225SE +/- 0.02, N = 3SE +/- 0.01, N = 34.794.82MIN: 4.7 / MAX: 5.85MIN: 4.78 / MAX: 5.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2Linux 5.10.2Linux 5.11 Git246810Min: 4.76 / Avg: 4.79 / Max: 4.81Min: 4.81 / Avg: 4.82 / Max: 4.851. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.20.14650.2930.43950.5860.7325SE +/- 0.002, N = 3SE +/- 0.004, N = 30.6470.6511. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Write - Average LatencyLinux 5.11 GitLinux 5.10.2246810Min: 0.64 / Avg: 0.65 / Max: 0.65Min: 0.64 / Avg: 0.65 / Max: 0.661. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

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.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesLinux 5.11 GitLinux 5.10.23691215SE +/- 0.04, N = 3SE +/- 0.06, N = 310.0910.151. (CC) gcc options: -lm
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesLinux 5.11 GitLinux 5.10.23691215Min: 10.04 / Avg: 10.09 / Max: 10.17Min: 10.09 / Avg: 10.15 / Max: 10.271. (CC) gcc options: -lm

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read WriteLinux 5.11 GitLinux 5.10.230060090012001500SE +/- 4.81, N = 3SE +/- 8.68, N = 3154615371. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read WriteLinux 5.11 GitLinux 5.10.230060090012001500Min: 1540.04 / Avg: 1545.94 / Max: 1555.46Min: 1525.12 / Avg: 1537.35 / Max: 1554.131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Matrix MathLinux 5.10.2Linux 5.11 Git13K26K39K52K65KSE +/- 232.35, N = 3SE +/- 164.05, N = 361027.3160672.251. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Matrix MathLinux 5.10.2Linux 5.11 Git11K22K33K44K55KMin: 60605.03 / Avg: 61027.31 / Max: 61406.44Min: 60472.61 / Avg: 60672.25 / Max: 60997.561. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

Zstd Compression

This test measures the time needed to compress a sample file (an Ubuntu ISO) using Zstd compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 3Linux 5.10.2Linux 5.11 Git13002600390052006500SE +/- 39.47, N = 3SE +/- 12.44, N = 35951.65917.81. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 3Linux 5.10.2Linux 5.11 Git10002000300040005000Min: 5874.3 / Avg: 5951.57 / Max: 6004.2Min: 5897.1 / Avg: 5917.8 / Max: 5940.11. (CC) gcc options: -O3 -pthread -lz

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0Linux 5.10.2Linux 5.11 Git1.19482.38963.58444.77925.974SE +/- 0.03, N = 3SE +/- 0.04, N = 35.285.31MIN: 5.2 / MAX: 6.76MIN: 5.24 / MAX: 5.581. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0Linux 5.10.2Linux 5.11 Git246810Min: 5.23 / Avg: 5.28 / Max: 5.32Min: 5.27 / Avg: 5.31 / Max: 5.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Linux 5.10.2Linux 5.11 Git1020304050SE +/- 0.24, N = 3SE +/- 0.03, N = 342.8943.131. (CC) gcc options: -O2 -ldl -lz -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Linux 5.10.2Linux 5.11 Git918273645Min: 42.55 / Avg: 42.89 / Max: 43.36Min: 43.06 / Avg: 43.13 / Max: 43.171. (CC) gcc options: -O2 -ldl -lz -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazefaceLinux 5.10.2Linux 5.11 Git0.4140.8281.2421.6562.07SE +/- 0.03, N = 3SE +/- 0.05, N = 31.831.84MIN: 1.78 / MAX: 2.05MIN: 1.77 / MAX: 2.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: blazefaceLinux 5.10.2Linux 5.11 Git246810Min: 1.79 / Avg: 1.83 / Max: 1.88Min: 1.79 / Avg: 1.84 / Max: 1.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnetLinux 5.11 GitLinux 5.10.20.85731.71462.57193.42924.2865SE +/- 0.01, N = 3SE +/- 0.01, N = 33.793.81MIN: 3.76 / MAX: 4.2MIN: 3.76 / MAX: 4.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: mnasnetLinux 5.11 GitLinux 5.10.2246810Min: 3.78 / Avg: 3.79 / Max: 3.8Min: 3.8 / Avg: 3.81 / Max: 3.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: CrownLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.04, N = 3SE +/- 0.05, N = 319.8519.75MIN: 19.66 / MAX: 20.29MIN: 19.52 / MAX: 20.25
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: CrownLinux 5.11 GitLinux 5.10.2510152025Min: 19.8 / Avg: 19.85 / Max: 19.93Min: 19.67 / Avg: 19.75 / Max: 19.84

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 4096Linux 5.10.2Linux 5.11 Git400K800K1200K1600K2000KSE +/- 9177.40, N = 3SE +/- 20492.72, N = 318535411844806
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 4096Linux 5.10.2Linux 5.11 Git300K600K900K1200K1500KMin: 1835283 / Avg: 1853541.33 / Max: 1864298Min: 1814449 / Avg: 1844806.33 / Max: 1883832

Build2

This test profile measures the time to bootstrap/install the build2 C++ build toolchain from source. Build2 is a cross-platform build toolchain for C/C++ code and features Cargo-like features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To CompileLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.53, N = 3SE +/- 0.96, N = 390.6191.03
OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To CompileLinux 5.11 GitLinux 5.10.220406080100Min: 89.59 / Avg: 90.61 / Max: 91.37Min: 89.49 / Avg: 91.03 / Max: 92.78

Opus Codec Encoding

Opus is an open audio codec. Opus is a lossy audio compression format designed primarily for interactive real-time applications over the Internet. This test uses Opus-Tools and measures the time required to encode a WAV file to Opus. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeLinux 5.10.2Linux 5.11 Git246810SE +/- 0.068, N = 5SE +/- 0.058, N = 56.1066.1341. (CXX) g++ options: -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeLinux 5.10.2Linux 5.11 Git246810Min: 5.93 / Avg: 6.11 / Max: 6.24Min: 5.95 / Avg: 6.13 / Max: 6.241. (CXX) g++ options: -fvisibility=hidden -logg -lm

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM TriadLinux 5.11 GitLinux 5.10.20.49160.98321.47481.96642.458SE +/- 0.00009, N = 3SE +/- 0.00104, N = 32.185092.175121. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: EP-STREAM TriadLinux 5.11 GitLinux 5.10.2246810Min: 2.18 / Avg: 2.19 / Max: 2.19Min: 2.17 / Avg: 2.18 / Max: 2.181. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPLLinux 5.11 GitLinux 5.10.21020304050SE +/- 0.19, N = 3SE +/- 0.11, N = 345.0844.891. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-HPLLinux 5.11 GitLinux 5.10.2918273645Min: 44.75 / Avg: 45.08 / Max: 45.42Min: 44.67 / Avg: 44.89 / Max: 45.011. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v2Linux 5.11 GitLinux 5.10.21.08682.17363.26044.34725.434SE +/- 0.00, N = 3SE +/- 0.01, N = 34.814.83MIN: 4.77 / MAX: 4.96MIN: 4.78 / MAX: 5.771. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: shufflenet-v2Linux 5.11 GitLinux 5.10.2246810Min: 4.8 / Avg: 4.81 / Max: 4.81Min: 4.81 / Avg: 4.83 / Max: 4.861. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: WASM imageConvolute - Browser: FirefoxLinux 5.10.2Linux 5.11 Git612182430SE +/- 0.26, N = 4SE +/- 0.24, N = 624.124.21. firefox 84.0
OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: WASM imageConvolute - Browser: FirefoxLinux 5.10.2Linux 5.11 Git612182430Min: 23.5 / Avg: 24.1 / Max: 24.7Min: 23.5 / Avg: 24.2 / Max: 25.31. firefox 84.0

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 3840 x 2160Linux 5.11 GitLinux 5.10.21122334455SE +/- 0.35, N = 3SE +/- 0.22, N = 349.549.31. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 3840 x 2160Linux 5.11 GitLinux 5.10.21020304050Min: 48.9 / Avg: 49.47 / Max: 50.1Min: 49 / Avg: 49.27 / Max: 49.71. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionLinux 5.10.2Linux 5.11 Git0.58281.16561.74842.33122.914SE +/- 0.00, N = 3SE +/- 0.03, N = 32.582.59
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionLinux 5.10.2Linux 5.11 Git246810Min: 2.57 / Avg: 2.58 / Max: 2.58Min: 2.56 / Avg: 2.59 / Max: 2.65

Unigine Superposition

This test calculates the average frame-rate within the Superposition demo for the Unigine engine, released in 2017. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Ultra - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git714212835SE +/- 0.03, N = 328.128.0MAX: 32.9MAX: 32.7
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: Ultra - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git612182430Min: 28 / Avg: 28.07 / Max: 28.1

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian DragonLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.16, N = 3SE +/- 0.19, N = 320.8120.73MIN: 20.5 / MAX: 21.49MIN: 20.25 / MAX: 21.26
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian DragonLinux 5.11 GitLinux 5.10.2510152025Min: 20.63 / Avg: 20.81 / Max: 21.12Min: 20.39 / Avg: 20.73 / Max: 21.03

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLinux 5.10.2Linux 5.11 Git11002200330044005500SE +/- 75.15, N = 15SE +/- 77.94, N = 155228.285210.821. (CC) gcc options: -O3 -mavx2
OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLinux 5.10.2Linux 5.11 Git9001800270036004500Min: 4784.58 / Avg: 5228.28 / Max: 5707.56Min: 4814.19 / Avg: 5210.82 / Max: 5820.481. (CC) gcc options: -O3 -mavx2

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.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsLinux 5.11 GitLinux 5.10.20.29030.58060.87091.16121.4515SE +/- 0.00139, N = 3SE +/- 0.00245, N = 31.286121.29036
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsLinux 5.11 GitLinux 5.10.2246810Min: 1.28 / Avg: 1.29 / Max: 1.29Min: 1.29 / Avg: 1.29 / Max: 1.29

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.05, N = 3SE +/- 0.08, N = 312.1812.22MIN: 12.01 / MAX: 12.69MIN: 12.05 / MAX: 12.551. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenetLinux 5.10.2Linux 5.11 Git48121620Min: 12.1 / Avg: 12.18 / Max: 12.28Min: 12.12 / Avg: 12.22 / Max: 12.381. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MallocLinux 5.10.2Linux 5.11 Git40M80M120M160M200MSE +/- 551101.63, N = 3SE +/- 266217.26, N = 3205233129.84204569594.711. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MallocLinux 5.10.2Linux 5.11 Git40M80M120M160M200MMin: 204324919.26 / Avg: 205233129.84 / Max: 206228071.69Min: 204040057.69 / Avg: 204569594.71 / Max: 204882402.781. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiLinux 5.10.2Linux 5.11 Git150300450600750SE +/- 2.33, N = 36796771. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiLinux 5.10.2Linux 5.11 Git120240360480600Min: 677 / Avg: 679.33 / Max: 6841. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git1.08052.1613.24154.3225.4025SE +/- 0.01252, N = 3SE +/- 0.01970, N = 34.788214.80215MIN: 4.54MIN: 4.51. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPULinux 5.10.2Linux 5.11 Git246810Min: 4.77 / Avg: 4.79 / Max: 4.81Min: 4.77 / Avg: 4.8 / Max: 4.831. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssdLinux 5.11 GitLinux 5.10.248121620SE +/- 0.03, N = 3SE +/- 0.05, N = 314.1614.20MIN: 13.96 / MAX: 14.52MIN: 13.95 / MAX: 15.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: squeezenet_ssdLinux 5.11 GitLinux 5.10.248121620Min: 14.12 / Avg: 14.16 / Max: 14.23Min: 14.13 / Avg: 14.2 / Max: 14.291. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaLinux 5.11 GitLinux 5.10.21326395265SE +/- 0.03, N = 3SE +/- 0.18, N = 359.2859.44
OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaLinux 5.11 GitLinux 5.10.21224364860Min: 59.23 / Avg: 59.28 / Max: 59.33Min: 59.14 / Avg: 59.44 / Max: 59.77

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: Kraken - Browser: FirefoxLinux 5.10.2Linux 5.11 Git2004006008001000SE +/- 5.02, N = 3SE +/- 5.47, N = 3917.7920.11. firefox 84.0
OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: Kraken - Browser: FirefoxLinux 5.10.2Linux 5.11 Git160320480640800Min: 910.7 / Avg: 917.67 / Max: 927.4Min: 909.3 / Avg: 920.07 / Max: 927.11. firefox 84.0

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: OpenGL 3.x - Resolution: 1920 x 1080Linux 5.11 GitLinux 5.10.22004006008001000SE +/- 0.00, N = 3SE +/- 1.84, N = 3982.9980.41. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: OpenGL 3.x - Resolution: 1920 x 1080Linux 5.11 GitLinux 5.10.22004006008001000Min: 982.9 / Avg: 982.9 / Max: 982.9Min: 976.8 / Avg: 980.37 / Max: 982.91. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3Linux 5.11 GitLinux 5.10.20.91.82.73.64.5SE +/- 0.01, N = 3SE +/- 0.01, N = 33.994.00MIN: 3.95 / MAX: 6.5MIN: 3.95 / MAX: 5.331. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v3-v3 - Model: mobilenet-v3Linux 5.11 GitLinux 5.10.2246810Min: 3.97 / Avg: 3.99 / Max: 4Min: 3.99 / Avg: 4 / Max: 4.021. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Embree

Intel Embree is a collection of high-performance ray-tracing kernels for execution on CPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: CrownLinux 5.11 GitLinux 5.10.2510152025SE +/- 0.04, N = 3SE +/- 0.06, N = 319.5319.48MIN: 19.31 / MAX: 20.01MIN: 19.23 / MAX: 19.88
OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: CrownLinux 5.11 GitLinux 5.10.2510152025Min: 19.48 / Avg: 19.53 / Max: 19.62Min: 19.39 / Avg: 19.48 / Max: 19.6

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Memory CopyingLinux 5.11 GitLinux 5.10.230060090012001500SE +/- 2.40, N = 3SE +/- 1.83, N = 31615.601611.731. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Memory CopyingLinux 5.11 GitLinux 5.10.230060090012001500Min: 1612.36 / Avg: 1615.6 / Max: 1620.29Min: 1609.27 / Avg: 1611.73 / Max: 1615.321. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinLinux 5.11 GitLinux 5.10.23691215SE +/- 0.02, N = 3SE +/- 0.07, N = 311.0711.041. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinLinux 5.11 GitLinux 5.10.23691215Min: 11.04 / Avg: 11.07 / Max: 11.1Min: 10.9 / Avg: 11.04 / Max: 11.131. (CXX) g++ options: -O3 -pthread -lm

eSpeak-NG Speech Engine

This test times how long it takes the eSpeak speech synthesizer to read Project Gutenberg's The Outline of Science and output to a WAV file. This test profile is now tracking the eSpeak-NG version of eSpeak. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisLinux 5.10.2Linux 5.11 Git510152025SE +/- 0.13, N = 4SE +/- 0.13, N = 421.7721.821. (CC) gcc options: -O2 -std=c99
OpenBenchmarking.orgSeconds, Fewer Is BettereSpeak-NG Speech Engine 20200907Text-To-Speech SynthesisLinux 5.10.2Linux 5.11 Git510152025Min: 21.46 / Avg: 21.77 / Max: 22.06Min: 21.52 / Avg: 21.82 / Max: 22.111. (CC) gcc options: -O2 -std=c99

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MMAPLinux 5.10.2Linux 5.11 Git80160240320400SE +/- 1.25, N = 3SE +/- 0.38, N = 3374.66373.891. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MMAPLinux 5.10.2Linux 5.11 Git70140210280350Min: 373.41 / Avg: 374.66 / Max: 377.15Min: 373.41 / Avg: 373.89 / Max: 374.651. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-FfteLinux 5.11 GitLinux 5.10.2246810SE +/- 0.06723, N = 3SE +/- 0.07039, N = 36.035406.023351. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.5.0Test / Class: G-FfteLinux 5.11 GitLinux 5.10.2246810Min: 5.93 / Avg: 6.04 / Max: 6.16Min: 5.95 / Avg: 6.02 / Max: 6.161. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

GLmark2

This is a test of Linaro's glmark2 port, currently using the X11 OpenGL 2.0 target. GLmark2 is a basic OpenGL benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterGLmark2 2020.04Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git40080012001600200020492045

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b0Linux 5.11 GitLinux 5.10.21.19032.38063.57094.76125.9515SE +/- 0.01, N = 3SE +/- 0.01, N = 35.285.29MIN: 5.24 / MAX: 5.66MIN: 5.24 / MAX: 6.681. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: efficientnet-b0Linux 5.11 GitLinux 5.10.2246810Min: 5.26 / Avg: 5.28 / Max: 5.29Min: 5.28 / Avg: 5.29 / Max: 5.321. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16Linux 5.10.2Linux 5.11 Git1224364860SE +/- 0.03, N = 3SE +/- 0.05, N = 353.4353.53MIN: 52.81 / MAX: 60.42MIN: 52.97 / MAX: 54.21. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: vgg16Linux 5.10.2Linux 5.11 Git1122334455Min: 53.37 / Avg: 53.43 / Max: 53.48Min: 53.44 / Avg: 53.53 / Max: 53.591. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.02, N = 3SE +/- 0.02, N = 311.7511.77
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: NoLinux 5.10.2Linux 5.11 Git3691215Min: 11.73 / Avg: 11.75 / Max: 11.78Min: 11.75 / Avg: 11.77 / Max: 11.8

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git9001800270036004500SE +/- 1.86, N = 3SE +/- 1.33, N = 3409040841. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git7001400210028003500Min: 4086 / Avg: 4089.67 / Max: 4092Min: 4083 / Avg: 4084.33 / Max: 40871. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MEMFDLinux 5.11 GitLinux 5.10.22004006008001000SE +/- 1.70, N = 3SE +/- 4.30, N = 31107.951106.361. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MEMFDLinux 5.11 GitLinux 5.10.22004006008001000Min: 1106.16 / Avg: 1107.95 / Max: 1111.35Min: 1099.97 / Avg: 1106.36 / Max: 1114.531. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.2400800120016002000SE +/- 19.15, N = 15SE +/- 22.45, N = 31811.341813.94MIN: 1701.55MIN: 1762.31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.230060090012001500Min: 1714.9 / Avg: 1811.34 / Max: 1914.41Min: 1780.77 / Avg: 1813.94 / Max: 1856.731. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: OpenGL 3.x - Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git2004006008001000SE +/- 1.50, N = 3SE +/- 7.43, N = 3976.8975.41. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: OpenGL 3.x - Resolution: 2560 x 1440Linux 5.10.2Linux 5.11 Git2004006008001000Min: 973.8 / Avg: 976.8 / Max: 978.3Min: 963.4 / Avg: 975.4 / Max: 9891. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

RealSR-NCNN

RealSR-NCNN is an NCNN neural network implementation of the RealSR project and accelerated using the Vulkan API. RealSR is the Real-World Super Resolution via Kernel Estimation and Noise Injection. NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. This test profile times how long it takes to increase the resolution of a sample image by a scale of 4x with Vulkan. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesLinux 5.11 GitLinux 5.10.220406080100SE +/- 0.11, N = 3SE +/- 0.03, N = 379.9480.05
OpenBenchmarking.orgSeconds, Fewer Is BetterRealSR-NCNN 20200818Scale: 4x - TAA: YesLinux 5.11 GitLinux 5.10.21530456075Min: 79.72 / Avg: 79.94 / Max: 80.05Min: 79.99 / Avg: 80.05 / Max: 80.11

Unigine Superposition

This test calculates the average frame-rate within the Superposition demo for the Unigine engine, released in 2017. This engine is extremely demanding on the system's graphics card. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: High - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git1632486480SE +/- 0.03, N = 3SE +/- 0.03, N = 370.670.5MAX: 84.7MAX: 84.7
OpenBenchmarking.orgFrames Per Second, More Is BetterUnigine Superposition 1.0Resolution: 1920 x 1080 - Mode: Fullscreen - Quality: High - Renderer: OpenGLLinux 5.10.2Linux 5.11 Git1428425670Min: 70.5 / Avg: 70.57 / Max: 70.6Min: 70.5 / Avg: 70.53 / Max: 70.6

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the Intel oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.20.70641.41282.11922.82563.532SE +/- 0.01051, N = 3SE +/- 0.00298, N = 33.135253.13966MIN: 3.03MIN: 3.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPULinux 5.11 GitLinux 5.10.2246810Min: 3.12 / Avg: 3.14 / Max: 3.15Min: 3.13 / Avg: 3.14 / Max: 3.141. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CryptoLinux 5.11 GitLinux 5.10.28001600240032004000SE +/- 8.97, N = 3SE +/- 8.27, N = 33867.493862.631. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CryptoLinux 5.11 GitLinux 5.10.27001400210028003500Min: 3854.8 / Avg: 3867.49 / Max: 3884.81Min: 3849.1 / Avg: 3862.63 / Max: 3877.651. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git110K220K330K440K550KSE +/- 3909.23, N = 3SE +/- 1197.96, N = 35248405242151. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git90K180K270K360K450KMin: 519112.76 / Avg: 524839.59 / Max: 532312.65Min: 522331.11 / Avg: 524215.48 / Max: 526439.151. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedLinux 5.10.2Linux 5.11 Git3K6K9K12K15KSE +/- 1.09, N = 3SE +/- 22.21, N = 613236.413220.71. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 9 - Decompression SpeedLinux 5.10.2Linux 5.11 Git2K4K6K8K10KMin: 13234.8 / Avg: 13236.43 / Max: 13238.5Min: 13149.3 / Avg: 13220.67 / Max: 13263.61. (CC) gcc options: -O3

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Linux 5.11 GitLinux 5.10.21.11852.2373.35554.4745.5925SE +/- 0.00574, N = 3SE +/- 0.00425, N = 34.971094.965301. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1Linux 5.11 GitLinux 5.10.2246810Min: 4.96 / Avg: 4.97 / Max: 4.98Min: 4.96 / Avg: 4.97 / Max: 4.971. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -pthread -lmpi_cxx -lmpi

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2120K240K360K480K600KSE +/- 7055.34, N = 3SE +/- 4430.88, N = 95403335397781. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.290K180K270K360K450KMin: 527000 / Avg: 540333.33 / Max: 551000Min: 522000 / Avg: 539777.78 / Max: 5660001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000204020381. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000SE +/- 0.67, N = 3SE +/- 0.67, N = 3204120391. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000Min: 2040 / Avg: 2041.33 / Max: 2042Min: 2038 / Avg: 2038.67 / Max: 20401. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedLinux 5.11 GitLinux 5.10.23K6K9K12K15KSE +/- 32.86, N = 3SE +/- 44.60, N = 313578.013564.71. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Decompression SpeedLinux 5.11 GitLinux 5.10.22K4K6K8K10KMin: 13513 / Avg: 13577.97 / Max: 13619Min: 13477.1 / Avg: 13564.7 / Max: 13623.11. (CC) gcc options: -O3

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.25001000150020002500SE +/- 27.28, N = 3SE +/- 17.30, N = 9211121091. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000Min: 2060 / Avg: 2111.33 / Max: 2153Min: 2039 / Avg: 2109 / Max: 22121. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git70140210280350SE +/- 2.80, N = 3SE +/- 2.12, N = 3300.70300.41
OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git50100150200250Min: 295.42 / Avg: 300.7 / Max: 304.97Min: 296.21 / Avg: 300.41 / Max: 303.05

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkLinux 5.10.2Linux 5.11 Git0.28910.57820.86731.15641.4455SE +/- 0.000, N = 3SE +/- 0.001, N = 31.2851.2841. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.3Water BenchmarkLinux 5.10.2Linux 5.11 Git246810Min: 1.29 / Avg: 1.29 / Max: 1.29Min: 1.28 / Avg: 1.28 / Max: 1.291. (CXX) g++ options: -O3 -pthread -lrt -lpthread -lm

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18Linux 5.11 GitLinux 5.10.23691215SE +/- 0.01, N = 3SE +/- 0.01, N = 313.0413.05MIN: 12.96 / MAX: 13.24MIN: 12.95 / MAX: 14.921. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU - Model: resnet18Linux 5.11 GitLinux 5.10.248121620Min: 13.02 / Avg: 13.04 / Max: 13.05Min: 13.02 / Avg: 13.05 / Max: 13.071. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

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.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 8Linux 5.11 GitLinux 5.10.250K100K150K200K250KSE +/- 792.15, N = 3SE +/- 466.03, N = 3214357.78214200.22MIN: 208730 / MAX: 228900MIN: 209060 / MAX: 226750
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: UDP - Test: Bandwidth - Threads: 8Linux 5.11 GitLinux 5.10.240K80K120K160K200KMin: 212830.67 / Avg: 214357.78 / Max: 215486.67Min: 213272.67 / Avg: 214200.22 / Max: 214743.33

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.29001800270036004500SE +/- 0.58, N = 3SE +/- 2.33, N = 3408640831. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.27001400210028003500Min: 4085 / Avg: 4086 / Max: 4087Min: 4079 / Avg: 4083.33 / Max: 40871. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git9001800270036004500SE +/- 0.58, N = 3SE +/- 0.88, N = 3408640831. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git7001400210028003500Min: 4085 / Avg: 4086 / Max: 4087Min: 4082 / Avg: 4083.33 / Max: 40851. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: WASM collisionDetection - Browser: FirefoxLinux 5.11 GitLinux 5.10.260120180240300SE +/- 2.28, N = 10SE +/- 3.88, N = 3296.2296.41. firefox 84.0
OpenBenchmarking.orgms, Fewer Is BetterSeleniumBenchmark: WASM collisionDetection - Browser: FirefoxLinux 5.11 GitLinux 5.10.250100150200250Min: 284.9 / Avg: 296.19 / Max: 307.4Min: 288.6 / Avg: 296.37 / Max: 300.31. firefox 84.0

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Vector MathLinux 5.11 GitLinux 5.10.230K60K90K120K150KSE +/- 72.42, N = 3SE +/- 76.71, N = 3135361.90135272.191. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Vector MathLinux 5.11 GitLinux 5.10.220K40K60K80K100KMin: 135260.05 / Avg: 135361.9 / Max: 135502.01Min: 135158.96 / Avg: 135272.19 / Max: 135418.451. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.210002000300040005000SE +/- 1.00, N = 3477047671. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.28001600240032004000Min: 4766 / Avg: 4767 / Max: 47691. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedLinux 5.10.2Linux 5.11 Git3K6K9K12K15KSE +/- 43.27, N = 3SE +/- 10.13, N = 312158.2512151.511. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 1 - Compression SpeedLinux 5.10.2Linux 5.11 Git2K4K6K8K10KMin: 12074.34 / Avg: 12158.25 / Max: 12218.55Min: 12138.18 / Avg: 12151.51 / Max: 12171.391. (CC) gcc options: -O3

Numpy Benchmark

This is a test to obtain the general Numpy performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkLinux 5.10.2Linux 5.11 Git110220330440550SE +/- 6.40, N = 3SE +/- 3.53, N = 3508.53508.25
OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkLinux 5.10.2Linux 5.11 Git90180270360450Min: 496.74 / Avg: 508.53 / Max: 518.76Min: 501.59 / Avg: 508.25 / Max: 513.62

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashLinux 5.10.2Linux 5.11 Git1.2M2.4M3.6M4.8M6MSE +/- 5947.55, N = 3SE +/- 3392.36, N = 3543946054365821. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashLinux 5.10.2Linux 5.11 Git900K1800K2700K3600K4500KMin: 5430974 / Avg: 5439460.33 / Max: 5450922Min: 5432022 / Avg: 5436582.33 / Max: 54432131. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000SE +/- 0.67, N = 3203920381. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000Min: 2037 / Avg: 2038.33 / Max: 20391. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Write - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git400800120016002000203920381. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.29001800270036004500408640841. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.25001000150020002500SE +/- 0.67, N = 3238123801. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Random Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000Min: 2379 / Avg: 2379.67 / Max: 23811. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: OpenGL 3.x - Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git2004006008001000SE +/- 1.84, N = 3SE +/- 6.17, N = 3980.8980.41. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: OpenGL 3.x - Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git2004006008001000Min: 978.3 / Avg: 980.83 / Max: 984.4Min: 969.3 / Avg: 980.43 / Max: 990.61. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBLinux 5.11 GitLinux 5.10.2612182430SE +/- 0.00, N = 3SE +/- 0.00, N = 326.5726.571. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBLinux 5.11 GitLinux 5.10.2612182430Min: 26.57 / Avg: 26.57 / Max: 26.58Min: 26.56 / Avg: 26.57 / Max: 26.571. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: AtomicLinux 5.10.2Linux 5.11 Git40K80K120K160K200KSE +/- 513.23, N = 3SE +/- 191.87, N = 3202668.16202613.611. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: AtomicLinux 5.10.2Linux 5.11 Git40K80K120K160K200KMin: 201897.48 / Avg: 202668.16 / Max: 203640.66Min: 202319.56 / Avg: 202613.61 / Max: 202974.151. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc Qsort Data SortingLinux 5.11 GitLinux 5.10.24080120160200SE +/- 0.45, N = 3SE +/- 0.70, N = 3204.00203.951. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc Qsort Data SortingLinux 5.11 GitLinux 5.10.24080120160200Min: 203.1 / Avg: 204 / Max: 204.47Min: 202.63 / Avg: 203.95 / Max: 205.031. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.270K140K210K280K350KSE +/- 3343.03, N = 15SE +/- 2645.75, N = 33280673280001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.260K120K180K240K300KMin: 301000 / Avg: 328066.67 / Max: 352000Min: 324000 / Avg: 328000 / Max: 3330001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

HPC Challenge

HPC Challenge (HPCC) is a cluster-focused benchmark consisting of the HPL Linpack TPP benchmark, DGEMM, STREAM, PTRANS, RandomAccess, FFT, and communication bandwidth and latency. This HPC Challenge test profile attempts to ship with standard yet versatile configuration/input files though they can be modified. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-PtransLinux 5.11 GitLinux 5.10.20.66731.33462.00192.66923.3365SE +/- 0.00150, N = 3SE +/- 0.00244, N = 32.965992.965691. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3
OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.5.0Test / Class: G-PtransLinux 5.11 GitLinux 5.10.2246810Min: 2.96 / Avg: 2.97 / Max: 2.97Min: 2.96 / Avg: 2.97 / Max: 2.971. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -funroll-loops2. OpenBLAS + Open MPI 4.0.3

LZ4 Compression

This test measures the time needed to compress/decompress a sample file (an Ubuntu ISO) using LZ4 compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedLinux 5.10.2Linux 5.11 Git3K6K9K12K15KSE +/- 25.05, N = 3SE +/- 51.98, N = 413150.213149.01. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterLZ4 Compression 1.9.3Compression Level: 3 - Decompression SpeedLinux 5.10.2Linux 5.11 Git2K4K6K8K10KMin: 13100.2 / Avg: 13150.2 / Max: 13177.8Min: 13016.4 / Avg: 13149.03 / Max: 13236.81. (CC) gcc options: -O3

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSeleniumBenchmark: Maze Solver - Browser: FirefoxLinux 5.10.2Linux 5.11 Git0.96751.9352.90253.874.8375SE +/- 0.00, N = 3SE +/- 0.04, N = 64.34.31. firefox 84.0
OpenBenchmarking.orgSeconds, Fewer Is BetterSeleniumBenchmark: Maze Solver - Browser: FirefoxLinux 5.10.2Linux 5.11 Git246810Min: 4.3 / Avg: 4.3 / Max: 4.3Min: 4.2 / Avg: 4.32 / Max: 4.41. firefox 84.0

NCNN

NCNN is a high performance neural network inference framework optimized for mobile and other platforms developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2Linux 5.10.2Linux 5.11 Git0.94951.8992.84853.7984.7475SE +/- 0.01, N = 3SE +/- 0.02, N = 34.224.22MIN: 4.17 / MAX: 5.41MIN: 4.16 / MAX: 12.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: Vulkan GPU-v2-v2 - Model: mobilenet-v2Linux 5.10.2Linux 5.11 Git246810Min: 4.21 / Avg: 4.22 / Max: 4.23Min: 4.2 / Avg: 4.22 / Max: 4.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Semaphore Release And AcquireLinux 5.10.2Linux 5.11 Git246810SE +/- 0.02, N = 3SE +/- 0.03, N = 38.168.161. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Semaphore Release And AcquireLinux 5.10.2Linux 5.11 Git3691215Min: 8.13 / Avg: 8.16 / Max: 8.18Min: 8.13 / Avg: 8.16 / Max: 8.221. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.03, N = 3SE +/- 0.00, N = 314.814.81. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexLinux 5.10.2Linux 5.11 Git48121620Min: 14.8 / Avg: 14.83 / Max: 14.9Min: 14.8 / Avg: 14.8 / Max: 14.81. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Shared Mutex Lock SharedLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.07, N = 3SE +/- 0.03, N = 315.715.71. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Shared Mutex Lock SharedLinux 5.10.2Linux 5.11 Git48121620Min: 15.6 / Avg: 15.67 / Max: 15.8Min: 15.7 / Avg: 15.73 / Max: 15.81. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git0.02140.04280.06420.08560.107SE +/- 0.001, N = 3SE +/- 0.000, N = 30.0950.0951. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git12345Min: 0.09 / Avg: 0.1 / Max: 0.1Min: 0.1 / Avg: 0.1 / Max: 0.11. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Zstd Compression

This test measures the time needed to compress a sample file (an Ubuntu ISO) using Zstd compression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 19Linux 5.11 GitLinux 5.10.21020304050SE +/- 0.03, N = 3SE +/- 0.06, N = 345.245.21. (CC) gcc options: -O3 -pthread -lz
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.5Compression Level: 19Linux 5.11 GitLinux 5.10.2918273645Min: 45.2 / Avg: 45.23 / Max: 45.3Min: 45.1 / Avg: 45.2 / Max: 45.31. (CC) gcc options: -O3 -pthread -lz

simdjson

This is a benchmark of SIMDJSON, a high performance JSON parser. SIMDJSON aims to be the fastest JSON parser and is used by projects like Microsoft FishStore, Yandex ClickHouse, Shopify, and others. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserIDLinux 5.11 GitLinux 5.10.20.1980.3960.5940.7920.99SE +/- 0.01, N = 3SE +/- 0.00, N = 30.880.881. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: DistinctUserIDLinux 5.11 GitLinux 5.10.2246810Min: 0.87 / Avg: 0.88 / Max: 0.89Min: 0.87 / Avg: 0.88 / Max: 0.881. (CXX) g++ options: -O3 -pthread

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandomLinux 5.11 GitLinux 5.10.20.13280.26560.39840.53120.664SE +/- 0.00, N = 3SE +/- 0.01, N = 30.590.591. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.7.1Throughput Test: LargeRandomLinux 5.11 GitLinux 5.10.2246810Min: 0.59 / Avg: 0.59 / Max: 0.6Min: 0.58 / Avg: 0.59 / Max: 0.61. (CXX) g++ options: -O3 -pthread

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 10Linux 5.11 GitLinux 5.10.23691215SE +/- 0.00, N = 3SE +/- 0.00, N = 310.510.51. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 10Linux 5.11 GitLinux 5.10.23691215Min: 10.5 / Avg: 10.5 / Max: 10.5Min: 10.5 / Avg: 10.5 / Max: 10.51. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 5Linux 5.11 GitLinux 5.10.21.1792.3583.5374.7165.895SE +/- 0.00, N = 3SE +/- 0.00, N = 35.245.241. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 5Linux 5.11 GitLinux 5.10.2246810Min: 5.24 / Avg: 5.24 / Max: 5.24Min: 5.24 / Avg: 5.24 / Max: 5.241. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 1Linux 5.11 GitLinux 5.10.20.23630.47260.70890.94521.1815SE +/- 0.00, N = 3SE +/- 0.00, N = 31.051.051. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 30 Seconds - Test: UDP - Parallel: 1Linux 5.11 GitLinux 5.10.2246810Min: 1.05 / Avg: 1.05 / Max: 1.05Min: 1.05 / Avg: 1.05 / Max: 1.051. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

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.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.230060090012001500SE +/- 13.03, N = 15SE +/- 11.61, N = 3128112811. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.22004006008001000Min: 1177 / Avg: 1281.33 / Max: 1374Min: 1264 / Avg: 1280.67 / Max: 13031. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.11 GitLinux 5.10.20.33750.6751.01251.351.6875SE +/- 0.00, N = 3SE +/- 0.00, N = 31.51.51. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.11 GitLinux 5.10.2246810Min: 1.5 / Avg: 1.5 / Max: 1.5Min: 1.5 / Avg: 1.5 / Max: 1.51. (CXX) g++ options: -O3 -lsnappy -lpthread

Selenium

This test profile uses the Selenium WebDriver for running various browser benchmarks in different available web browsers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: MotionMark - Browser: FirefoxLinux 5.10.2Linux 5.11 Git90180270360450SE +/- 16.52, N = 9SE +/- 22.70, N = 9399.96362.231. firefox 84.0
OpenBenchmarking.orgScore, More Is BetterSeleniumBenchmark: MotionMark - Browser: FirefoxLinux 5.10.2Linux 5.11 Git70140210280350Min: 304.27 / Avg: 399.96 / Max: 459.94Min: 272.11 / Avg: 362.23 / Max: 482.411. firefox 84.0

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git0.00970.01940.02910.03880.0485SE +/- 0.001, N = 3SE +/- 0.003, N = 150.0320.0431. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git12345Min: 0.03 / Avg: 0.03 / Max: 0.03Min: 0.03 / Avg: 0.04 / Max: 0.061. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git7K14K21K28K35KSE +/- 302.18, N = 3SE +/- 1640.45, N = 1531316245341. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git5K10K15K20K25KMin: 30737.81 / Avg: 31315.73 / Max: 31757.78Min: 17143.73 / Avg: 24534.46 / Max: 33182.261. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git0.00810.01620.02430.03240.0405SE +/- 0.000, N = 3SE +/- 0.002, N = 150.0280.0361. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average LatencyLinux 5.10.2Linux 5.11 Git12345Min: 0.03 / Avg: 0.03 / Max: 0.03Min: 0.03 / Avg: 0.04 / Max: 0.051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git8K16K24K32K40KSE +/- 487.24, N = 3SE +/- 1834.50, N = 1536550293311. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read OnlyLinux 5.10.2Linux 5.11 Git6K12K18K24K30KMin: 35877.05 / Avg: 36549.88 / Max: 37496.77Min: 19975.13 / Avg: 29330.95 / Max: 38313.861. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Intel Open Image Denoise

Open Image Denoise is a denoising library for ray-tracing and part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.00, N = 3SE +/- 0.35, N = 1211.9110.15
OpenBenchmarking.orgImages / Sec, More Is BetterIntel Open Image Denoise 1.2.0Scene: MemorialLinux 5.10.2Linux 5.11 Git3691215Min: 11.9 / Avg: 11.91 / Max: 11.91Min: 7.63 / Avg: 10.15 / Max: 11.65

CLOMP

CLOMP is the C version of the Livermore OpenMP benchmark developed to measure OpenMP overheads and other performance impacts due to threading in order to influence future system designs. This particular test profile configuration is currently set to look at the OpenMP static schedule speed-up across all available CPU cores using the recommended test configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP SpeedupLinux 5.11 GitLinux 5.10.248121620SE +/- 0.10, N = 3SE +/- 0.28, N = 1517.717.21. (CC) gcc options: -fopenmp -O3 -lm
OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP SpeedupLinux 5.11 GitLinux 5.10.248121620Min: 17.6 / Avg: 17.7 / Max: 17.9Min: 14.6 / Avg: 17.21 / Max: 17.91. (CC) gcc options: -fopenmp -O3 -lm

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 4096Linux 5.10.2Linux 5.11 Git300K600K900K1200K1500KSE +/- 2999.73, N = 3SE +/- 26718.42, N = 1212904341289524
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 4096Linux 5.10.2Linux 5.11 Git200K400K600K800K1000KMin: 1284579 / Avg: 1290434.33 / Max: 1294494Min: 999663 / Avg: 1289524.17 / Max: 1353852

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.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesLinux 5.11 GitLinux 5.10.2612182430SE +/- 0.88, N = 15SE +/- 1.45, N = 1221.5323.731. (CC) gcc options: -lm
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesLinux 5.11 GitLinux 5.10.2612182430Min: 18.23 / Avg: 21.53 / Max: 27.53Min: 19.12 / Avg: 23.73 / Max: 30.731. (CC) gcc options: -lm

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBLinux 5.10.2Linux 5.11 Git816243240SE +/- 1.54, N = 12SE +/- 0.20, N = 332.7022.901. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma
OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBLinux 5.10.2Linux 5.11 Git714212835Min: 23.32 / Avg: 32.7 / Max: 41.9Min: 22.62 / Avg: 22.9 / Max: 23.291. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lslang -lz -llzma -lnuma

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.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 16Linux 5.11 GitLinux 5.10.220K40K60K80K100KSE +/- 10244.72, N = 15SE +/- 11316.74, N = 1310234173000
OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 16Linux 5.11 GitLinux 5.10.220K40K60K80K100KMin: 29570 / Avg: 102340.67 / Max: 156150Min: 5490 / Avg: 73000 / Max: 136510

OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 8Linux 5.11 GitLinux 5.10.212K24K36K48K60KSE +/- 1166.03, N = 15SE +/- 1144.33, N = 125772455999
OpenBenchmarking.orgConnections/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Connections/s - Threads: 8Linux 5.11 GitLinux 5.10.210K20K30K40K50KMin: 52880 / Avg: 57724 / Max: 65820Min: 51650 / Avg: 55999.17 / Max: 64890

OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1Linux 5.10.2Linux 5.11 Git48121620SE +/- 0.36, N = 12SE +/- 0.20, N = 1517.4718.00MIN: 6.25 / MAX: 24.09MIN: 7.07 / MAX: 24.26
OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1Linux 5.10.2Linux 5.11 Git510152025Min: 15.33 / Avg: 17.47 / Max: 20.03Min: 16.57 / Avg: 18 / Max: 19.09

yquake2

This is a test of Yamagi Quake II. Yamagi Quake II is an enhanced client for id Software's Quake II with focus on offline and coop gameplay. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git4080120160200SE +/- 0.85, N = 3SE +/- 11.05, N = 15179.1152.91. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC
OpenBenchmarking.orgFrames Per Second, More Is Betteryquake2 7.45Renderer: Software CPU - Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git306090120150Min: 178.1 / Avg: 179.1 / Max: 180.8Min: 89.3 / Avg: 152.86 / Max: 190.61. (CC) gcc options: -lm -ldl -rdynamic -shared -lSDL2 -O2 -pipe -fomit-frame-pointer -std=gnu99 -fno-strict-aliasing -fwrapv -fvisibility=hidden -MMD -mfpmath=sse -fPIC

Xonotic

This is a benchmark of Xonotic, which is a fork of the DarkPlaces-based Nexuiz game. Development began in March of 2010 on the Xonotic game. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltimateLinux 5.10.2Linux 5.11 Git70140210280350SE +/- 0.42, N = 3SE +/- 4.77, N = 15329.77239.83MIN: 63 / MAX: 541MIN: 56 / MAX: 543
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: UltimateLinux 5.10.2Linux 5.11 Git60120180240300Min: 328.98 / Avg: 329.77 / Max: 330.41Min: 221.83 / Avg: 239.83 / Max: 276.08

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: HighLinux 5.10.2Linux 5.11 Git100200300400500SE +/- 0.74, N = 3SE +/- 25.17, N = 12471.83328.09MIN: 190 / MAX: 698MIN: 165 / MAX: 697
OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.8.2Resolution: 3840 x 2160 - Effects Quality: HighLinux 5.10.2Linux 5.11 Git80160240320400Min: 470.81 / Avg: 471.83 / Max: 473.26Min: 263.84 / Avg: 328.09 / Max: 470.99

Tesseract

Tesseract is a fork of Cube 2 Sauerbraten with numerous graphics and game-play improvements. Tesseract has been in development since 2012 while its first release happened in May of 2014. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git110220330440550SE +/- 6.45, N = 15SE +/- 12.58, N = 15488.29399.10
OpenBenchmarking.orgFrames Per Second, More Is BetterTesseract 2014-05-12Resolution: 1920 x 1080Linux 5.10.2Linux 5.11 Git90180270360450Min: 454.96 / Avg: 488.29 / Max: 551.07Min: 352.76 / Avg: 399.1 / Max: 502.81

ET: Legacy

ETLegacy is an open-source engine evolution of Wolfenstein: Enemy Territory, a World War II era first person shooter that was released for free by Splash Damage using the id Tech 3 engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.75Renderer: Renderer2 - Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git60120180240300SE +/- 1.23, N = 3SE +/- 8.51, N = 15278.4182.0
OpenBenchmarking.orgFrames Per Second, More Is BetterET: Legacy 2.75Renderer: Renderer2 - Resolution: 3840 x 2160Linux 5.10.2Linux 5.11 Git50100150200250Min: 276.5 / Avg: 278.37 / Max: 280.7Min: 154.9 / Avg: 182.03 / Max: 276.4

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.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git150K300K450K600K750KSE +/- 42763.95, N = 12SE +/- 8569.97, N = 37141677053331. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git120K240K360K480K600KMin: 247000 / Avg: 714166.67 / Max: 789000Min: 689000 / Avg: 705333.33 / Max: 7180001. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git6001200180024003000SE +/- 167.02, N = 12SE +/- 33.26, N = 3279027561. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git5001000150020002500Min: 965 / Avg: 2789.67 / Max: 3081Min: 2693 / Avg: 2756 / Max: 28061. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git30060090012001500SE +/- 183.92, N = 12SE +/- 138.35, N = 15156913111. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git30060090012001500Min: 991 / Avg: 1568.5 / Max: 2325Min: 985 / Avg: 1310.8 / Max: 23561. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git7001400210028003500SE +/- 367.90, N = 12SE +/- 276.68, N = 15314526291. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: Linux AIO - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.10.2Linux 5.11 Git5001000150020002500Min: 1990 / Avg: 3144.58 / Max: 4658Min: 1977 / Avg: 2628.67 / Max: 47181. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.2400800120016002000SE +/- 129.36, N = 12SE +/- 118.43, N = 15195219311. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.230060090012001500Min: 1145 / Avg: 1951.58 / Max: 2384Min: 1043 / Avg: 1930.93 / Max: 23811. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.28001600240032004000SE +/- 258.72, N = 12SE +/- 236.96, N = 15391138691. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.25Type: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11 GitLinux 5.10.27001400210028003500Min: 2298 / Avg: 3910.58 / Max: 4776Min: 2093 / Avg: 3869.2 / Max: 47701. (CC) gcc options: -rdynamic -ll -lnuma -lrt -lz -lpthread -lm -ldl -laio -lcurl -lssl -lcrypto -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomLinux 5.10.2Linux 5.11 Git48121620SE +/- 0.03, N = 3SE +/- 0.38, N = 1514.1016.431. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomLinux 5.10.2Linux 5.11 Git48121620Min: 14.05 / Avg: 14.1 / Max: 14.14Min: 14.41 / Avg: 16.43 / Max: 20.611. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.062, N = 3SE +/- 0.342, N = 159.99612.0081. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadLinux 5.10.2Linux 5.11 Git3691215Min: 9.9 / Avg: 10 / Max: 10.11Min: 10.62 / Avg: 12.01 / Max: 15.411. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadLinux 5.10.2Linux 5.11 Git3691215SE +/- 0.015, N = 3SE +/- 0.243, N = 159.87511.6751. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadLinux 5.10.2Linux 5.11 Git3691215Min: 9.85 / Avg: 9.88 / Max: 9.9Min: 10.19 / Avg: 11.68 / Max: 13.71. (CXX) g++ options: -O3 -lsnappy -lpthread

280 Results Shown

Xonotic
IPC_benchmark
Ethr
Stress-NG
oneDNN
Tesseract
Selenium
Stress-NG
Ethr
Stress-NG
MPV
Selenium:
  StyleBench - Firefox
  Jetstream - Firefox
HPC Challenge
Ethr
OpenVKL
Unigine Valley
perf-bench
Selenium
PostgreSQL pgbench
Selenium
iPerf
PostgreSQL pgbench
Unigine Heaven
InfluxDB
Selenium
MPV
oneDNN
Ethr
IPC_benchmark
Hackbench:
  32 - Process
  16 - Process
Ethr
Hackbench:
  1 - Process
  8 - Process
  1 - Thread
  16 - Thread
VkFFT
oneDNN
IPC_benchmark
Hackbench
Flexible IO Tester:
  Rand Read - Linux AIO - No - Yes - 4KB - Default Test Directory:
    MB/s
    IOPS
InfluxDB
Unigine Superposition
LevelDB:
  Overwrite:
    MB/s
    Microseconds Per Op
oneDNN
Hackbench
Flexible IO Tester:
  Rand Write - Linux AIO - No - Yes - 4KB - Default Test Directory:
    IOPS
    MB/s
Unigine Valley
Stress-NG
iPerf
LevelDB
Stress-NG
BenchmarkMutex
Ogg Audio Encoding
LevelDB
Node.js V8 Web Tooling Benchmark
Hackbench
NCNN
Flexible IO Tester:
  Rand Write - IO_uring - No - Yes - 4KB - Default Test Directory:
    MB/s
    IOPS
NCNN
OSBench
IPC_benchmark
HPC Challenge
KeyDB
LevelDB
Stress-NG
Timed Eigen Compilation
NCNN
Facebook RocksDB
GLmark2
IPC_benchmark
Stress-NG
Selenium
NCNN
Facebook RocksDB
LevelDB
NCNN
Facebook RocksDB
OSBench
IPC_benchmark
OSBench
Selenium
LZ4 Compression
LevelDB
iPerf
LZ4 Compression
IPC_benchmark
LevelDB
Facebook RocksDB
Flexible IO Tester
BenchmarkMutex
Flexible IO Tester
IPC_benchmark
Ethr
NCNN
IPC_benchmark
NCNN
Stress-NG
NCNN:
  CPU-v3-v3 - mobilenet-v3
  CPU - squeezenet_ssd
BenchmarkMutex
VKMark
Xonotic
oneDNN
Hackbench
NCNN
WavPack Audio Encoding
GLmark2
perf-bench
Mlpack Benchmark
BenchmarkMutex
NCNN:
  CPU-v2-v2 - mobilenet-v2
  Vulkan GPU - regnety_400m
Stress-NG
Embree
NCNN
Facebook RocksDB
Waifu2x-NCNN Vulkan
perf-bench
NCNN
Ethr
VKMark
HPC Challenge
Flexible IO Tester:
  Rand Read - Linux AIO - No - Yes - 2MB - Default Test Directory:
    IOPS
    MB/s
Timed Linux Kernel Compilation
IPC_benchmark
yquake2
Timed HMMer Search
Ethr
NCNN
Timed LLVM Compilation
NCNN
PostgreSQL pgbench
HPC Challenge
PostgreSQL pgbench
NCNN
Mlpack Benchmark
PostgreSQL pgbench
Flexible IO Tester
NCNN
PostgreSQL pgbench
VKMark
PostgreSQL pgbench
Flexible IO Tester
BenchmarkMutex
Ethr
HPC Challenge
Unigine Heaven
PostgreSQL pgbench
Hackbench
Stress-NG
NCNN
PostgreSQL pgbench
OSBench
PostgreSQL pgbench
Stress-NG
Zstd Compression
NCNN
SQLite Speedtest
NCNN:
  Vulkan GPU - blazeface
  Vulkan GPU - mnasnet
Embree
IPC_benchmark
Build2
Opus Codec Encoding
HPC Challenge:
  EP-STREAM Triad
  G-HPL
NCNN
Selenium
yquake2
Mlpack Benchmark
Unigine Superposition
Embree
Himeno Benchmark
NAMD
NCNN
Stress-NG
perf-bench
oneDNN
NCNN
Mlpack Benchmark
Selenium
yquake2
NCNN
Embree
Stress-NG
LAMMPS Molecular Dynamics Simulator
eSpeak-NG Speech Engine
Stress-NG
HPC Challenge
GLmark2
NCNN:
  Vulkan GPU - efficientnet-b0
  Vulkan GPU - vgg16
RealSR-NCNN
Flexible IO Tester
Stress-NG
oneDNN
yquake2
RealSR-NCNN
Unigine Superposition
oneDNN
Stress-NG
PostgreSQL pgbench
LZ4 Compression
High Performance Conjugate Gradient
Flexible IO Tester:
  Seq Write - Linux AIO - No - Yes - 4KB - Default Test Directory
  Seq Write - Linux AIO - No - Yes - 2MB - Default Test Directory
  Rand Write - IO_uring - No - Yes - 2MB - Default Test Directory
LZ4 Compression
Flexible IO Tester
Tesseract
GROMACS
NCNN
Ethr
Flexible IO Tester:
  Seq Write - Linux AIO - No - Yes - 2MB - Default Test Directory
  Rand Write - Linux AIO - No - Yes - 2MB - Default Test Directory
Selenium
Stress-NG
Flexible IO Tester
LZ4 Compression
Numpy Benchmark
perf-bench
Flexible IO Tester:
  Seq Write - IO_uring - No - Yes - 2MB - Default Test Directory
  Rand Write - Linux AIO - No - Yes - 2MB - Default Test Directory
  Seq Write - IO_uring - No - Yes - 2MB - Default Test Directory
  Rand Read - IO_uring - No - Yes - 2MB - Default Test Directory
yquake2
perf-bench
Stress-NG:
  Atomic
  Glibc Qsort Data Sorting
Flexible IO Tester
HPC Challenge
LZ4 Compression
Selenium
NCNN
BenchmarkMutex:
  Semaphore Release And Acquire
  Mutex Lock Unlock std::mutex
  Shared Mutex Lock Shared
PostgreSQL pgbench
Zstd Compression
simdjson:
  DistinctUserID
  LargeRand
iPerf:
  5201 - 30 Seconds - UDP - 10
  5201 - 30 Seconds - UDP - 5
  5201 - 30 Seconds - UDP - 1
Flexible IO Tester
LevelDB
Selenium
PostgreSQL pgbench:
  100 - 1 - Read Only - Average Latency
  100 - 1 - Read Only
  1 - 1 - Read Only - Average Latency
  1 - 1 - Read Only
Intel Open Image Denoise
CLOMP
IPC_benchmark
OSBench
perf-bench
Ethr:
  TCP - Connections/s - 16
  TCP - Connections/s - 8
  TCP - Latency - 1
yquake2
Xonotic:
  3840 x 2160 - Ultimate
  3840 x 2160 - High
Tesseract
ET: Legacy
Flexible IO Tester:
  Seq Write - IO_uring - No - Yes - 4KB - Default Test Directory:
    IOPS
    MB/s
  Seq Read - Linux AIO - No - Yes - 2MB - Default Test Directory:
    IOPS
    MB/s
  Seq Read - IO_uring - No - Yes - 2MB - Default Test Directory:
    IOPS
    MB/s
LevelDB:
  Seek Rand
  Rand Read
  Hot Read

Linux 5.10.2

Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (2702 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211

OS: Ubuntu 20.10, Kernel: 5.10.2-051002-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
Graphics Notes: GLAMOR
Python Notes: Python 3.8.6
Security 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 December 2020 10:32 by user phoronix.

Linux 5.11 Git

Processor: AMD Ryzen 9 5900X 12-Core @ 3.70GHz (12 Cores / 24 Threads), Motherboard: ASUS ROG CROSSHAIR VIII HERO (2702 BIOS), Chipset: AMD Starship/Matisse, Memory: 16GB, Disk: 1000GB Sabrent Rocket 4.0 1TB, Graphics: Sapphire AMD Radeon RX 5600 OEM/5600 XT / 5700/5700 6GB (1780/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel I211

OS: Ubuntu 20.10, Kernel: 5.10.0-051000daily20201222-generic (x86_64), Desktop: GNOME Shell 3.38.1, Display Server: X Server 1.20.9, Display Driver: modesetting 1.20.9, OpenGL: 4.6 Mesa 20.2.1 (LLVM 11.0.0), Vulkan: 1.2.131, Compiler: GCC 10.2.0, File-System: ext4, Screen Resolution: 3840x2160

Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-10-JvwpWM/gcc-10-10.2.0/debian/tmp-gcn/usr,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 -v
Disk Notes: NONE / errors=remount-ro,relatime,rw / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq schedutil (Boost: Enabled) - CPU Microcode: 0xa201009
Graphics Notes: GLAMOR
Python Notes: Python 3.8.6
Security 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 December 2020 05:56 by user phoronix.