Core i9 10980XE Linux 5.12

Intel Core i9-10980XE testing with a ASRock X299 Steel Legend (P1.30 BIOS) and llvmpipe on Ubuntu 20.04 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 2103053-PTS-COREI91080
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:

C++ Boost Tests 2 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 6 Tests
CPU Massive 10 Tests
Creator Workloads 2 Tests
Database Test Suite 5 Tests
Disk Test Suite 3 Tests
HPC - High Performance Computing 4 Tests
Common Kernel Benchmarks 6 Tests
Molecular Dynamics 3 Tests
Multi-Core 8 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 5 Tests
Python Tests 2 Tests
Scientific Computing 3 Tests
Server 6 Tests
Server CPU Tests 9 Tests
Single-Threaded 2 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
Performance Per
Dollar
Date
Run
  Test
  Duration
Linux 5.11.3
March 05 2021
  5 Hours, 20 Minutes
Linux 5.12 Git
March 05 2021
  5 Hours, 8 Minutes
Invert Hiding All Results Option
  5 Hours, 14 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 5.11.3

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: llvmpipe, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211

OS: Ubuntu 20.04, Kernel: 5.11.3-051103-generic (x86_64), Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 3.3 Mesa 20.0.8 (LLVM 10.0.0 256 bits), Compiler: GCC 9.3.0 + CUDA 11.1, File-System: ext4, Screen Resolution: 1024x768

Kernel Notes: Transparent Huge Pages: madvise
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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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: intel_cpufreq ondemand - CPU Microcode: 0x5003003
Python Notes: Python 2.7.18 + Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Linux 5.12 Git

OS: Ubuntu 20.04, Kernel: 5.12.0-051200rc1daily20210305-generic (x86_64) 20210304, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 3.3 Mesa 20.0.8 (LLVM 10.0.0 256 bits), Compiler: GCC 9.3.0 + CUDA 11.1, File-System: ext4, Screen Resolution: 1024x768

Core i9 10980XE Linux 5.12OpenBenchmarking.orgPhoronix Test SuiteIntel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads)ASRock X299 Steel Legend (P1.30 BIOS)Intel Sky Lake-E DMI3 Registers32GBSamsung SSD 970 PRO 512GBllvmpipeRealtek ALC1220Intel I219-V + Intel I211Ubuntu 20.045.11.3-051103-generic (x86_64)5.12.0-051200rc1daily20210305-generic (x86_64) 20210304GNOME Shell 3.36.4X Server 1.20.93.3 Mesa 20.0.8 (LLVM 10.0.0 256 bits)GCC 9.3.0 + CUDA 11.1ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelsDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionCore I9 10980XE Linux 5.12 BenchmarksSystem Logs- Transparent Huge Pages: madvise- --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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: intel_cpufreq ondemand - CPU Microcode: 0x5003003- Python 2.7.18 + Python 3.8.5- itlb_multihit: KVM: Mitigation of VMX disabled + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Linux 5.11.3 vs. Linux 5.12 Git ComparisonPhoronix Test SuiteBaseline+14.7%+14.7%+29.4%+29.4%+44.1%+44.1%4.1%3.7%3.3%2.8%2.5%2.2%2.1%Context Switching58.6%256MB7.1%64MB5.3%NUMA4.4%512MB100 - 100 - Read Write - Average Latency3.8%100 - 100 - Read Write3.8%S.V.M.P1.F.1.S3.5%CompileSENDFILE3.2%Rand Fill Sync3%Read While Writing2.9%Hot Read100 - 50 - Read Only2.7%ForkingRand Write - IO_uring - No - Yes - 4KB2.5%Rand Write - IO_uring - No - Yes - 4KB2.4%Seq Write - IO_uring - No - Yes - 4KB2.4%Rand Delete2.4%Seq Write - IO_uring - No - Yes - 4KB2.4%Atomic32 - Process5.F.1.S.4.T2.1%Overwrite2.1%Fill Sync2.1%Stress-NGIORIORStress-NGIORPostgreSQL pgbenchPostgreSQL pgbenchStress-NGFS-MarkCompile BenchStress-NGFacebook RocksDBFacebook RocksDBLevelDBPostgreSQL pgbenchStress-NGFlexible IO TesterFlexible IO TesterFlexible IO TesterLevelDBFlexible IO TesterStress-NGHackbenchFS-MarkLevelDBLevelDBLinux 5.11.3Linux 5.12 Git

Core i9 10980XE Linux 5.12ior: 256MB - Default Test Directoryior: 64MB - Default Test Directorystress-ng: NUMAior: 512MB - Default Test Directorypgbench: 100 - 100 - Read Write - Average Latencypgbench: 100 - 100 - Read Writestress-ng: System V Message Passingfs-mark: 1000 Files, 1MB Sizecompilebench: Compilestress-ng: SENDFILErocksdb: Rand Fill Syncrocksdb: Read While Writingleveldb: Hot Readpgbench: 100 - 50 - Read Onlystress-ng: Forkingfio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Rand Write - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Write - IO_uring - No - Yes - 4KB - Default Test Directoryleveldb: Rand Deletefio: Seq Write - IO_uring - No - Yes - 4KB - Default Test Directorystress-ng: Atomichackbench: 32 - Processfs-mark: 5000 Files, 1MB Size, 4 Threadsleveldb: Overwriteleveldb: Fill Syncleveldb: Rand Fillkeydb: compress-zstd: 19 - Decompression Speedpgbench: 100 - 50 - Read Only - Average Latencystress-ng: MEMFDleveldb: Seq Fillfs-mark: 4000 Files, 32 Sub Dirs, 1MB Sizecompilebench: Initial Createleveldb: Seq Fillleveldb: Rand Fillleveldb: Overwritecompress-zstd: 19, Long Mode - Compression Speedcompilebench: Read Compiled Treengspice: C2670rocksdb: Rand Fillcompress-zstd: 19, Long Mode - Decompression Speedrocksdb: Seq Filldav1d: Summer Nature 4Kleveldb: Rand Readstress-ng: CPU Cachepgbench: 100 - 100 - Read Only - Average Latencystress-ng: Socket Activitypgbench: 100 - 100 - Read Onlypgbench: 100 - 50 - Read Writebuild-wasmer: Time To Compilepgbench: 100 - 50 - Read Write - Average Latencycompress-zstd: 19 - Compression Speedctx-clock: Context Switch Timeopenfoam: Motorbike 30Mfio: Seq Read - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Read - IO_uring - No - Yes - 4KB - Default Test Directoryngspice: C7552influxdb: 1024 - 10000 - 2,5000,1 - 10000compress-zstd: 8, Long Mode - Compression Speedfio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directorybuild-linux-kernel: Time To Compilestress-ng: Memory Copyingleveldb: Seek Randstress-ng: CPU Stressstress-ng: Glibc Qsort Data Sortingtensorflow-lite: NASNet Mobilerocksdb: Rand Readfio: Seq Write - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Read - IO_uring - No - Yes - 4KB - Default Test Directoryfio: Seq Write - IO_uring - No - Yes - 2MB - Default Test Directoryinfluxdb: 64 - 10000 - 2,5000,1 - 10000build-erlang: Time To Compilegromacs: water_GMX50_barestress-ng: Glibc C String Functionsphpbench: PHP Benchmark Suitetensorflow-lite: SqueezeNetnamd: ATPase Simulation - 327,506 Atomsstress-ng: Semaphoresfio: Seq Read - IO_uring - No - Yes - 2MB - Default Test Directorycompress-zstd: 8, Long Mode - Decompression Speedquantlib: stress-ng: Cryptofio: Seq Read - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Read - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Read - IO_uring - No - Yes - 2MB - Default Test Directorytensorflow-lite: Mobilenet Quantstress-ng: Matrix Mathdav1d: Summer Nature 1080pinfluxdb: 4 - 10000 - 2,5000,1 - 10000tensorflow-lite: Inception V4stress-ng: Mallocstress-ng: MMAPtensorflow-lite: Mobilenet Floatstress-ng: Vector Mathstress-ng: RdRandtensorflow-lite: Inception ResNet V2pybench: Total For Average Test Timesfio: Rand Write - IO_uring - No - Yes - 2MB - Default Test Directoryfio: Rand Write - IO_uring - No - Yes - 2MB - Default Test Directoryleveldb: Fill Syncstress-ng: Context SwitchingLinux 5.11.3Linux 5.12 Git1091.051113.05401.941198.248.638115918368739.11168.42346.76346212.434673438115028.52145973481169.664716671842518000318.0722022161924.4160.813460.9325.1907480.366325.501693242.713189.80.1091074.42331.168167.0688.6412.012.212.243.73141.10150.86613756183211.41517377236.5628.22692.350.2089942.15482051861273.6875.80860.1129120.221895485000140.9251624941.9533.434066753.7522570.6633.5297241.39279.81118093935630951118132922421601095.8119.1421.5731113785.378049351280510.978612536812.9931313793.92540.63711.5715621562313290133.895276.20585.811241189.11856223186144479.08458.4787370.2131734.70184099.861641737912111722410.56860046.151018.361056.88385.111247.698.964111728676595.64162.72425.27335412.514539425963227.75444772883207.204603331798505667325.6801975165455.5459.557451.4332.0127636.556331.664706276.033132.30.1111056.00336.880164.2700.3311.812.01243.03096.13148.70813949433167.11538240239.3527.92191.370.21010037.18477836853973.0695.85759.6128119.311884482333141.6301631665.9535.633933353.5422579.7033.6467216.39278.94118453938177221115132622371597752.9119.3731.5761112027.208038411282110.977472534325.8631343791.12538.93709.1815631563313490076.795335.43585.471240590.91855363186058640.44458.5287377.4131729.34184094.091641773912111722410.54326056.01OpenBenchmarking.org

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 256MB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git2004006008001000SE +/- 4.25, N = 3SE +/- 4.85, N = 31091.051018.36MIN: 811.86 / MAX: 1238.26MIN: 528.45 / MAX: 1212.361. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 256MB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git2004006008001000Min: 1082.59 / Avg: 1091.05 / Max: 1095.96Min: 1008.69 / Avg: 1018.36 / Max: 1023.831. (CC) gcc options: -O2 -lm -pthread -lmpi

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 64MB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git2004006008001000SE +/- 5.34, N = 3SE +/- 4.15, N = 31113.051056.88MIN: 772.08 / MAX: 1469.24MIN: 377.95 / MAX: 1430.161. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 64MB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git2004006008001000Min: 1104.01 / Avg: 1113.05 / Max: 1122.49Min: 1048.99 / Avg: 1056.88 / Max: 1063.081. (CC) gcc options: -O2 -lm -pthread -lmpi

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 5.11.3Linux 5.12 Git90180270360450SE +/- 3.88, N = 3SE +/- 0.83, N = 3401.94385.111. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: NUMALinux 5.11.3Linux 5.12 Git70140210280350Min: 395.88 / Avg: 401.94 / Max: 409.16Min: 383.46 / Avg: 385.11 / Max: 386.111. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

IOR

IOR is a parallel I/O storage benchmark making use of MPI with a particular focus on HPC (High Performance Computing) systems. IOR is developed at the Lawrence Livermore National Laboratory (LLNL). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 512MB - Disk Target: Default Test DirectoryLinux 5.12 GitLinux 5.11.330060090012001500SE +/- 3.35, N = 3SE +/- 2.20, N = 31247.691198.24MIN: 859.46 / MAX: 1391.34MIN: 777.16 / MAX: 1382.81. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.3.0Block Size: 512MB - Disk Target: Default Test DirectoryLinux 5.12 GitLinux 5.11.32004006008001000Min: 1244.21 / Avg: 1247.69 / Max: 1254.38Min: 1193.88 / Avg: 1198.24 / Max: 1200.971. (CC) gcc options: -O2 -lm -pthread -lmpi

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: 100 - Mode: Read Write - Average LatencyLinux 5.11.3Linux 5.12 Git3691215SE +/- 0.034, N = 3SE +/- 0.049, N = 38.6388.9641. (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: 100 - Mode: Read Write - Average LatencyLinux 5.11.3Linux 5.12 Git3691215Min: 8.6 / Avg: 8.64 / Max: 8.71Min: 8.87 / Avg: 8.96 / Max: 9.021. (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: 100 - Mode: Read WriteLinux 5.11.3Linux 5.12 Git2K4K6K8K10KSE +/- 46.17, N = 3SE +/- 61.28, N = 311591111721. (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: 100 - Mode: Read WriteLinux 5.11.3Linux 5.12 Git2K4K6K8K10KMin: 11498.69 / Avg: 11590.64 / Max: 11643.95Min: 11097.8 / Avg: 11172.28 / Max: 11293.81. (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: System V Message PassingLinux 5.12 GitLinux 5.11.32M4M6M8M10MSE +/- 22579.96, N = 3SE +/- 21475.99, N = 38676595.648368739.111. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: System V Message PassingLinux 5.12 GitLinux 5.11.31.5M3M4.5M6M7.5MMin: 8638709.75 / Avg: 8676595.64 / Max: 8716823.74Min: 8325814.45 / Avg: 8368739.11 / Max: 8391527.631. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

FS-Mark

FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 1000 Files, 1MB SizeLinux 5.11.3Linux 5.12 Git4080120160200SE +/- 0.27, N = 3SE +/- 0.78, N = 3168.4162.71. (CC) gcc options: -static
OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 1000 Files, 1MB SizeLinux 5.11.3Linux 5.12 Git306090120150Min: 168.1 / Avg: 168.37 / Max: 168.9Min: 161.4 / Avg: 162.7 / Max: 164.11. (CC) gcc options: -static

Compile Bench

Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompileLinux 5.12 GitLinux 5.11.35001000150020002500SE +/- 5.25, N = 3SE +/- 9.25, N = 32425.272346.76
OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompileLinux 5.12 GitLinux 5.11.3400800120016002000Min: 2415.87 / Avg: 2425.27 / Max: 2434.01Min: 2334.51 / Avg: 2346.76 / Max: 2364.9

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 5.11.3Linux 5.12 Git70K140K210K280K350KSE +/- 170.46, N = 3SE +/- 119.12, N = 3346212.43335412.511. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SENDFILELinux 5.11.3Linux 5.12 Git60K120K180K240K300KMin: 345871.94 / Avg: 346212.43 / Max: 346397.52Min: 335187.59 / Avg: 335412.51 / Max: 335593.011. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncLinux 5.11.3Linux 5.12 Git10002000300040005000SE +/- 19.94, N = 3SE +/- 10.59, N = 3467345391. (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 5.11.3Linux 5.12 Git8001600240032004000Min: 4635 / Avg: 4673.33 / Max: 4702Min: 4518 / Avg: 4538.67 / Max: 45531. (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 5.11.3Linux 5.12 Git900K1800K2700K3600K4500KSE +/- 37311.60, N = 15SE +/- 44123.79, N = 8438115042596321. (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 5.11.3Linux 5.12 Git800K1600K2400K3200K4000KMin: 4210941 / Avg: 4381149.67 / Max: 4663607Min: 4166097 / Avg: 4259631.88 / Max: 45172771. (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: Hot ReadLinux 5.12 GitLinux 5.11.3714212835SE +/- 0.14, N = 3SE +/- 0.16, N = 327.7528.521. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadLinux 5.12 GitLinux 5.11.3612182430Min: 27.55 / Avg: 27.75 / Max: 28.01Min: 28.2 / Avg: 28.52 / Max: 28.721. (CXX) g++ options: -O3 -lsnappy -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: 100 - Clients: 50 - Mode: Read OnlyLinux 5.11.3Linux 5.12 Git100K200K300K400K500KSE +/- 2026.55, N = 3SE +/- 1044.75, N = 34597344477281. (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 5.11.3Linux 5.12 Git80K160K240K320K400KMin: 456799.62 / Avg: 459733.8 / Max: 463622.38Min: 445645.27 / Avg: 447728.02 / Max: 448914.731. (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: ForkingLinux 5.12 GitLinux 5.11.320K40K60K80K100KSE +/- 304.86, N = 3SE +/- 187.97, N = 383207.2081169.661. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: ForkingLinux 5.12 GitLinux 5.11.314K28K42K56K70KMin: 82765.91 / Avg: 83207.2 / Max: 83792.22Min: 80873.57 / Avg: 81169.66 / Max: 81518.321. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -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: Random Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git100K200K300K400K500KSE +/- 2666.67, N = 3SE +/- 2666.67, N = 34716674603331. (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 5.11.3Linux 5.12 Git80K160K240K320K400KMin: 469000 / Avg: 471666.67 / Max: 477000Min: 455000 / Avg: 460333.33 / Max: 4630001. (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 5.11.3Linux 5.12 Git400800120016002000SE +/- 10.87, N = 3SE +/- 9.68, N = 3184217981. (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 5.11.3Linux 5.12 Git30060090012001500Min: 1830 / Avg: 1842.33 / Max: 1864Min: 1779 / Avg: 1798.33 / Max: 18091. (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 5.11.3Linux 5.12 Git110K220K330K440K550KSE +/- 1732.05, N = 3SE +/- 1452.97, N = 35180005056671. (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 5.11.3Linux 5.12 Git90K180K270K360K450KMin: 515000 / Avg: 518000 / Max: 521000Min: 503000 / Avg: 505666.67 / Max: 5080001. (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: Random DeleteLinux 5.11.3Linux 5.12 Git70140210280350SE +/- 0.27, N = 3SE +/- 0.73, N = 3318.07325.681. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteLinux 5.11.3Linux 5.12 Git60120180240300Min: 317.56 / Avg: 318.07 / Max: 318.5Min: 324.3 / Avg: 325.68 / Max: 326.781. (CXX) g++ options: -O3 -lsnappy -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: Sequential Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git400800120016002000SE +/- 6.64, N = 3SE +/- 5.78, N = 3202219751. (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 5.11.3Linux 5.12 Git400800120016002000Min: 2011 / Avg: 2022.33 / Max: 2034Min: 1965 / Avg: 1975.33 / Max: 19851. (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: AtomicLinux 5.12 GitLinux 5.11.340K80K120K160K200KSE +/- 1670.69, N = 3SE +/- 2271.88, N = 4165455.54161924.411. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: AtomicLinux 5.12 GitLinux 5.11.330K60K90K120K150KMin: 162114.56 / Avg: 165455.54 / Max: 167170.79Min: 156049.11 / Avg: 161924.41 / Max: 166070.621. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

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 5.12 GitLinux 5.11.31428425670SE +/- 0.49, N = 3SE +/- 0.39, N = 359.5660.811. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessLinux 5.12 GitLinux 5.11.31224364860Min: 58.57 / Avg: 59.56 / Max: 60.09Min: 60.04 / Avg: 60.81 / Max: 61.211. (CC) gcc options: -lpthread

FS-Mark

FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 5000 Files, 1MB Size, 4 ThreadsLinux 5.11.3Linux 5.12 Git100200300400500SE +/- 0.35, N = 3SE +/- 2.91, N = 3460.9451.41. (CC) gcc options: -static
OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 5000 Files, 1MB Size, 4 ThreadsLinux 5.11.3Linux 5.12 Git80160240320400Min: 460.4 / Avg: 460.93 / Max: 461.6Min: 445.6 / Avg: 451.4 / Max: 454.81. (CC) gcc options: -static

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: OverwriteLinux 5.11.3Linux 5.12 Git70140210280350SE +/- 0.40, N = 3SE +/- 0.54, N = 3325.19332.011. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11.3Linux 5.12 Git60120180240300Min: 324.45 / Avg: 325.19 / Max: 325.8Min: 330.98 / Avg: 332.01 / Max: 332.821. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.11.3Linux 5.12 Git16003200480064008000SE +/- 11.82, N = 3SE +/- 45.68, N = 37480.377636.561. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.11.3Linux 5.12 Git13002600390052006500Min: 7465.59 / Avg: 7480.37 / Max: 7503.73Min: 7590.14 / Avg: 7636.56 / Max: 7727.911. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11.3Linux 5.12 Git70140210280350SE +/- 1.08, N = 3SE +/- 0.86, N = 3325.50331.661. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11.3Linux 5.12 Git60120180240300Min: 324.3 / Avg: 325.5 / Max: 327.66Min: 330.33 / Avg: 331.66 / Max: 333.271. (CXX) g++ options: -O3 -lsnappy -lpthread

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 5.12 GitLinux 5.11.3150K300K450K600K750KSE +/- 1560.70, N = 3SE +/- 3003.18, N = 3706276.03693242.711. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16Linux 5.12 GitLinux 5.11.3120K240K360K480K600KMin: 704578.92 / Avg: 706276.03 / Max: 709393.34Min: 689854.23 / Avg: 693242.71 / Max: 699231.811. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedLinux 5.11.3Linux 5.12 Git7001400210028003500SE +/- 13.64, N = 3SE +/- 17.48, N = 33189.83132.31. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Decompression SpeedLinux 5.11.3Linux 5.12 Git6001200180024003000Min: 3164.7 / Avg: 3189.8 / Max: 3211.6Min: 3112.3 / Avg: 3132.27 / Max: 3167.11. (CC) gcc options: -O3 -pthread -lz -llzma

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 5.11.3Linux 5.12 Git0.0250.050.0750.10.125SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1090.1111. (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 5.11.3Linux 5.12 Git12345Min: 0.11 / Avg: 0.11 / Max: 0.11Min: 0.11 / Avg: 0.11 / Max: 0.111. (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: MEMFDLinux 5.11.3Linux 5.12 Git2004006008001000SE +/- 0.22, N = 3SE +/- 1.83, N = 31074.421056.001. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MEMFDLinux 5.11.3Linux 5.12 Git2004006008001000Min: 1073.98 / Avg: 1074.42 / Max: 1074.67Min: 1054.08 / Avg: 1056 / Max: 1059.651. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

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 5.11.3Linux 5.12 Git70140210280350SE +/- 1.06, N = 3SE +/- 0.14, N = 3331.17336.881. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillLinux 5.11.3Linux 5.12 Git60120180240300Min: 329.8 / Avg: 331.17 / Max: 333.26Min: 336.62 / Avg: 336.88 / Max: 337.11. (CXX) g++ options: -O3 -lsnappy -lpthread

FS-Mark

FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 4000 Files, 32 Sub Dirs, 1MB SizeLinux 5.11.3Linux 5.12 Git4080120160200SE +/- 0.20, N = 3SE +/- 0.17, N = 3167.0164.21. (CC) gcc options: -static
OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 4000 Files, 32 Sub Dirs, 1MB SizeLinux 5.11.3Linux 5.12 Git306090120150Min: 166.8 / Avg: 167 / Max: 167.4Min: 163.9 / Avg: 164.23 / Max: 164.41. (CC) gcc options: -static

Compile Bench

Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial CreateLinux 5.12 GitLinux 5.11.3150300450600750SE +/- 5.11, N = 3SE +/- 3.18, N = 3700.33688.64
OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial CreateLinux 5.12 GitLinux 5.11.3120240360480600Min: 690.84 / Avg: 700.33 / Max: 708.37Min: 682.41 / Avg: 688.64 / Max: 692.81

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 5.11.3Linux 5.12 Git3691215SE +/- 0.06, N = 3SE +/- 0.00, N = 312.011.81. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillLinux 5.11.3Linux 5.12 Git3691215Min: 11.9 / Avg: 12 / Max: 12.1Min: 11.8 / Avg: 11.8 / Max: 11.81. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11.3Linux 5.12 Git3691215SE +/- 0.07, N = 3SE +/- 0.03, N = 312.212.01. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillLinux 5.11.3Linux 5.12 Git48121620Min: 12.1 / Avg: 12.23 / Max: 12.3Min: 11.9 / Avg: 11.97 / Max: 121. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11.3Linux 5.12 Git3691215SE +/- 0.03, N = 312.212.01. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteLinux 5.11.3Linux 5.12 Git48121620Min: 12.2 / Avg: 12.23 / Max: 12.31. (CXX) g++ options: -O3 -lsnappy -lpthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression SpeedLinux 5.11.3Linux 5.12 Git1020304050SE +/- 0.07, N = 3SE +/- 0.25, N = 343.743.01. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Compression SpeedLinux 5.11.3Linux 5.12 Git918273645Min: 43.6 / Avg: 43.73 / Max: 43.8Min: 42.7 / Avg: 43 / Max: 43.51. (CC) gcc options: -O3 -pthread -lz -llzma

Compile Bench

Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Read Compiled TreeLinux 5.11.3Linux 5.12 Git7001400210028003500SE +/- 44.32, N = 3SE +/- 30.16, N = 33141.103096.13
OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Read Compiled TreeLinux 5.11.3Linux 5.12 Git5001000150020002500Min: 3081.7 / Avg: 3141.1 / Max: 3227.78Min: 3046.86 / Avg: 3096.13 / Max: 3150.89

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Linux 5.12 GitLinux 5.11.3306090120150SE +/- 0.54, N = 3SE +/- 0.81, N = 3148.71150.871. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C2670Linux 5.12 GitLinux 5.11.3306090120150Min: 147.94 / Avg: 148.71 / Max: 149.76Min: 149.25 / Avg: 150.87 / Max: 151.831. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

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 5.12 GitLinux 5.11.3300K600K900K1200K1500KSE +/- 14664.33, N = 3SE +/- 1691.99, N = 3139494313756181. (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 5.12 GitLinux 5.11.3200K400K600K800K1000KMin: 1365721 / Avg: 1394942.67 / Max: 1411721Min: 1372392 / Avg: 1375618 / Max: 13781161. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedLinux 5.11.3Linux 5.12 Git7001400210028003500SE +/- 13.07, N = 3SE +/- 56.37, N = 33211.43167.11. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19, Long Mode - Decompression SpeedLinux 5.11.3Linux 5.12 Git6001200180024003000Min: 3186.8 / Avg: 3211.43 / Max: 3231.3Min: 3055.1 / Avg: 3167.13 / Max: 3234.11. (CC) gcc options: -O3 -pthread -lz -llzma

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 5.12 GitLinux 5.11.3300K600K900K1200K1500KSE +/- 13798.52, N = 3SE +/- 4462.40, N = 3153824015173771. (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 5.12 GitLinux 5.11.3300K600K900K1200K1500KMin: 1511580 / Avg: 1538240 / Max: 1557745Min: 1509593 / Avg: 1517377.33 / Max: 15250501. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 4KLinux 5.12 GitLinux 5.11.350100150200250SE +/- 0.49, N = 3SE +/- 0.45, N = 3239.35236.56MIN: 183.7 / MAX: 263.89MIN: 184.72 / MAX: 260.051. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 4KLinux 5.12 GitLinux 5.11.34080120160200Min: 238.47 / Avg: 239.35 / Max: 240.15Min: 235.77 / Avg: 236.56 / Max: 237.331. (CC) gcc options: -pthread -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 ReadLinux 5.12 GitLinux 5.11.3714212835SE +/- 0.14, N = 3SE +/- 0.22, N = 327.9228.231. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadLinux 5.12 GitLinux 5.11.3612182430Min: 27.76 / Avg: 27.92 / Max: 28.2Min: 27.99 / Avg: 28.23 / Max: 28.661. (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: CPU CacheLinux 5.11.3Linux 5.12 Git20406080100SE +/- 0.97, N = 15SE +/- 1.15, N = 592.3591.371. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU CacheLinux 5.11.3Linux 5.12 Git20406080100Min: 85.16 / Avg: 92.35 / Max: 98Min: 88.03 / Avg: 91.37 / Max: 94.461. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -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.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average LatencyLinux 5.11.3Linux 5.12 Git0.04730.09460.14190.18920.2365SE +/- 0.000, N = 3SE +/- 0.001, N = 30.2080.2101. (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: 100 - Mode: Read Only - Average LatencyLinux 5.11.3Linux 5.12 Git12345Min: 0.21 / Avg: 0.21 / Max: 0.21Min: 0.21 / Avg: 0.21 / Max: 0.211. (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: Socket ActivityLinux 5.12 GitLinux 5.11.32K4K6K8K10KSE +/- 74.76, N = 3SE +/- 70.08, N = 310037.189942.151. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Socket ActivityLinux 5.12 GitLinux 5.11.32K4K6K8K10KMin: 9888.11 / Avg: 10037.18 / Max: 10121.68Min: 9802.78 / Avg: 9942.15 / Max: 10024.681. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -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: 100 - Clients: 100 - Mode: Read OnlyLinux 5.11.3Linux 5.12 Git100K200K300K400K500KSE +/- 1052.76, N = 3SE +/- 3384.33, N = 34820514778361. (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: 100 - Mode: Read OnlyLinux 5.11.3Linux 5.12 Git80K160K240K320K400KMin: 480119.91 / Avg: 482050.79 / Max: 483743.35Min: 474377.32 / Avg: 477835.88 / Max: 4846041. (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 5.11.3Linux 5.12 Git2K4K6K8K10KSE +/- 10.80, N = 3SE +/- 19.20, N = 3861285391. (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 5.11.3Linux 5.12 Git15003000450060007500Min: 8598.24 / Avg: 8611.96 / Max: 8633.26Min: 8506.37 / Avg: 8538.85 / Max: 8572.831. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Timed Wasmer Compilation

This test times how long it takes to compile Wasmer. Wasmer is written in the Rust programming language and is a WebAssembly runtime implementation that supports WASI and EmScripten. This test profile builds Wasmer with the Cranelift and Singlepast compiler features enabled. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileLinux 5.12 GitLinux 5.11.31632486480SE +/- 0.07, N = 3SE +/- 0.17, N = 373.0773.691. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileLinux 5.12 GitLinux 5.11.31428425670Min: 72.93 / Avg: 73.07 / Max: 73.17Min: 73.42 / Avg: 73.69 / Max: 741. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil

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 5.11.3Linux 5.12 Git1.31782.63563.95345.27126.589SE +/- 0.007, N = 3SE +/- 0.013, N = 35.8085.8571. (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 5.11.3Linux 5.12 Git246810Min: 5.79 / Avg: 5.81 / Max: 5.82Min: 5.83 / Avg: 5.86 / Max: 5.881. (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/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression SpeedLinux 5.11.3Linux 5.12 Git1326395265SE +/- 0.38, N = 3SE +/- 0.49, N = 360.159.61. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 19 - Compression SpeedLinux 5.11.3Linux 5.12 Git1224364860Min: 59.4 / Avg: 60.13 / Max: 60.7Min: 58.6 / Avg: 59.57 / Max: 60.21. (CC) gcc options: -O3 -pthread -lz -llzma

ctx_clock

Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch TimeLinux 5.12 GitLinux 5.11.3306090120150SE +/- 0.67, N = 3128129
OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch TimeLinux 5.12 GitLinux 5.11.320406080100Min: 127 / Avg: 128.33 / Max: 129

OpenFOAM

OpenFOAM is the leading free, open source software for computational fluid dynamics (CFD). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MLinux 5.12 GitLinux 5.11.3306090120150SE +/- 0.65, N = 3SE +/- 0.55, N = 3119.31120.221. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -ldl -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MLinux 5.12 GitLinux 5.11.320406080100Min: 118.3 / Avg: 119.31 / Max: 120.51Min: 119.16 / Avg: 120.22 / Max: 121.021. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfoamToVTK -ldynamicMesh -llagrangian -lgenericPatchFields -lfileFormats -lOpenFOAM -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: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 4KB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git400800120016002000SE +/- 4.98, N = 3SE +/- 2.19, N = 3189518841. (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 5.11.3Linux 5.12 Git30060090012001500Min: 1887 / Avg: 1894.67 / Max: 1904Min: 1881 / Avg: 1883.67 / Max: 18881. (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 5.11.3Linux 5.12 Git100K200K300K400K500KSE +/- 333.33, N = 34850004823331. (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 5.11.3Linux 5.12 Git80K160K240K320K400KMin: 482000 / Avg: 482333.33 / Max: 4830001. (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

Ngspice

Ngspice is an open-source SPICE circuit simulator. Ngspice was originally based on the Berkeley SPICE electronic circuit simulator. Ngspice supports basic threading using OpenMP. This test profile is making use of the ISCAS 85 benchmark circuits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Linux 5.11.3Linux 5.12 Git306090120150SE +/- 1.60, N = 3SE +/- 1.41, N = 3140.93141.631. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE
OpenBenchmarking.orgSeconds, Fewer Is BetterNgspice 34Circuit: C7552Linux 5.11.3Linux 5.12 Git306090120150Min: 137.73 / Avg: 140.93 / Max: 142.56Min: 138.81 / Avg: 141.63 / Max: 143.121. (CC) gcc options: -O0 -fopenmp -lm -lstdc++ -lfftw3 -lXaw -lXmu -lXt -lXext -lX11 -lSM -lICE

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: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.12 GitLinux 5.11.3300K600K900K1200K1500KSE +/- 2085.74, N = 3SE +/- 2969.34, N = 31631665.91624941.9
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 1024 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.12 GitLinux 5.11.3300K600K900K1200K1500KMin: 1628230.4 / Avg: 1631665.93 / Max: 1635432.8Min: 1620427 / Avg: 1624941.9 / Max: 1630540.4

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Compression SpeedLinux 5.12 GitLinux 5.11.3120240360480600SE +/- 2.97, N = 3SE +/- 2.08, N = 3535.6533.41. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Compression SpeedLinux 5.12 GitLinux 5.11.390180270360450Min: 531.6 / Avg: 535.6 / Max: 541.4Min: 529.8 / Avg: 533.43 / Max: 5371. (CC) gcc options: -O3 -pthread -lz -llzma

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 5.11.3Linux 5.12 Git70K140K210K280K350KSE +/- 333.33, N = 3SE +/- 333.33, N = 33406673393331. (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 5.11.3Linux 5.12 Git60K120K180K240K300KMin: 340000 / Avg: 340666.67 / Max: 341000Min: 339000 / Avg: 339333.33 / Max: 3400001. (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 (defconfig) for the architecture being tested. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileLinux 5.12 GitLinux 5.11.31224364860SE +/- 0.71, N = 4SE +/- 0.93, N = 353.5453.75
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileLinux 5.12 GitLinux 5.11.31122334455Min: 52.7 / Avg: 53.54 / Max: 55.67Min: 52.8 / Avg: 53.75 / Max: 55.61

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 5.12 GitLinux 5.11.36001200180024003000SE +/- 3.46, N = 3SE +/- 2.78, N = 32579.702570.661. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Memory CopyingLinux 5.12 GitLinux 5.11.3400800120016002000Min: 2573.38 / Avg: 2579.7 / Max: 2585.3Min: 2565.24 / Avg: 2570.66 / Max: 2574.461. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

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 5.11.3Linux 5.12 Git816243240SE +/- 0.08, N = 3SE +/- 0.20, N = 333.5333.651. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomLinux 5.11.3Linux 5.12 Git714212835Min: 33.37 / Avg: 33.53 / Max: 33.65Min: 33.27 / Avg: 33.65 / Max: 33.941. (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: CPU StressLinux 5.11.3Linux 5.12 Git16003200480064008000SE +/- 10.59, N = 3SE +/- 1.88, N = 37241.397216.391. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU StressLinux 5.11.3Linux 5.12 Git13002600390052006500Min: 7222.19 / Avg: 7241.39 / Max: 7258.75Min: 7213.24 / Avg: 7216.39 / Max: 7219.751. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc Qsort Data SortingLinux 5.11.3Linux 5.12 Git60120180240300SE +/- 0.34, N = 3SE +/- 0.55, N = 3279.81278.941. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc Qsort Data SortingLinux 5.11.3Linux 5.12 Git50100150200250Min: 279.17 / Avg: 279.81 / Max: 280.33Min: 277.87 / Avg: 278.94 / Max: 279.671. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileLinux 5.11.3Linux 5.12 Git30K60K90K120K150KSE +/- 601.90, N = 3SE +/- 317.38, N = 3118093118453
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileLinux 5.11.3Linux 5.12 Git20K40K60K80K100KMin: 117293 / Avg: 118093 / Max: 119272Min: 117844 / Avg: 118453.33 / Max: 118912

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 5.12 GitLinux 5.11.320M40M60M80M100MSE +/- 66112.60, N = 3SE +/- 138364.67, N = 393817722935630951. (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 5.12 GitLinux 5.11.316M32M48M64M80MMin: 93690451 / Avg: 93817722 / Max: 93912409Min: 93302831 / Avg: 93563095 / Max: 937746601. (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 Write - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.11.3Linux 5.12 Git2004006008001000SE +/- 0.33, N = 3111811151. (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 5.11.3Linux 5.12 Git2004006008001000Min: 1114 / Avg: 1114.67 / Max: 11151. (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 5.11.3Linux 5.12 Git30060090012001500SE +/- 1.53, N = 3SE +/- 1.20, N = 3132913261. (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 5.11.3Linux 5.12 Git2004006008001000Min: 1326 / Avg: 1329 / Max: 1331Min: 1324 / Avg: 1325.67 / Max: 13281. (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 5.11.3Linux 5.12 Git5001000150020002500SE +/- 0.67, N = 3SE +/- 0.88, N = 3224222371. (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 5.11.3Linux 5.12 Git400800120016002000Min: 2241 / Avg: 2242.33 / Max: 2243Min: 2235 / Avg: 2236.67 / Max: 22381. (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: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.11.3Linux 5.12 Git300K600K900K1200K1500KSE +/- 1998.89, N = 3SE +/- 835.73, N = 31601095.81597752.9
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.11.3Linux 5.12 Git300K600K900K1200K1500KMin: 1597126.5 / Avg: 1601095.77 / Max: 1603493.2Min: 1596696.3 / Avg: 1597752.87 / Max: 1599402.8

Timed Erlang/OTP Compilation

This test times how long it takes to compile Erlang/OTP. Erlang is a programming language and run-time for massively scalable soft real-time systems with high availability requirements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To CompileLinux 5.11.3Linux 5.12 Git306090120150SE +/- 0.12, N = 3SE +/- 0.10, N = 3119.14119.37
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To CompileLinux 5.11.3Linux 5.12 Git20406080100Min: 118.97 / Avg: 119.14 / Max: 119.38Min: 119.17 / Avg: 119.37 / Max: 119.49

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 2021Input: water_GMX50_bareLinux 5.12 GitLinux 5.11.30.35460.70921.06381.41841.773SE +/- 0.000, N = 3SE +/- 0.001, N = 31.5761.5731. (CXX) g++ options: -O3 -pthread
OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2021Input: water_GMX50_bareLinux 5.12 GitLinux 5.11.3246810Min: 1.58 / Avg: 1.58 / Max: 1.58Min: 1.57 / Avg: 1.57 / Max: 1.571. (CXX) g++ options: -O3 -pthread

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 5.11.3Linux 5.12 Git200K400K600K800K1000KSE +/- 5947.71, N = 3SE +/- 919.70, N = 31113785.371112027.201. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc C String FunctionsLinux 5.11.3Linux 5.12 Git200K400K600K800K1000KMin: 1102095.88 / Avg: 1113785.37 / Max: 1121538.64Min: 1110824.32 / Avg: 1112027.2 / Max: 1113833.781. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

PHPBench

PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteLinux 5.11.3Linux 5.12 Git200K400K600K800K1000KSE +/- 1912.09, N = 3SE +/- 3829.08, N = 3804935803841
OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteLinux 5.11.3Linux 5.12 Git140K280K420K560K700KMin: 801155 / Avg: 804935 / Max: 807327Min: 796282 / Avg: 803841.33 / Max: 808683

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetLinux 5.11.3Linux 5.12 Git30K60K90K120K150KSE +/- 26.54, N = 3SE +/- 36.95, N = 3128051128211
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetLinux 5.11.3Linux 5.12 Git20K40K60K80K100KMin: 127998 / Avg: 128051 / Max: 128080Min: 128147 / Avg: 128210.67 / Max: 128275

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 5.12 GitLinux 5.11.30.22020.44040.66060.88081.101SE +/- 0.00563, N = 3SE +/- 0.00290, N = 30.977470.97861
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsLinux 5.12 GitLinux 5.11.3246810Min: 0.97 / Avg: 0.98 / Max: 0.99Min: 0.97 / Avg: 0.98 / Max: 0.98

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 5.11.3Linux 5.12 Git500K1000K1500K2000K2500KSE +/- 1174.79, N = 3SE +/- 1085.56, N = 32536812.992534325.861. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SemaphoresLinux 5.11.3Linux 5.12 Git400K800K1200K1600K2000KMin: 2534493.66 / Avg: 2536812.99 / Max: 2538298.1Min: 2532641.95 / Avg: 2534325.86 / Max: 2536354.671. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -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: Sequential Read - Engine: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.12 GitLinux 5.11.37001400210028003500313431311. (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

Zstd Compression

This test measures the time needed to compress/decompress a sample file (a FreeBSD disk image - FreeBSD-12.2-RELEASE-amd64-memstick.img) using Zstd compression with options for different compression levels / settings. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Decompression SpeedLinux 5.11.3Linux 5.12 Git8001600240032004000SE +/- 7.76, N = 3SE +/- 12.11, N = 33793.93791.11. (CC) gcc options: -O3 -pthread -lz -llzma
OpenBenchmarking.orgMB/s, More Is BetterZstd Compression 1.4.9Compression Level: 8, Long Mode - Decompression SpeedLinux 5.11.3Linux 5.12 Git7001400210028003500Min: 3782.9 / Avg: 3793.93 / Max: 3808.9Min: 3768.9 / Avg: 3791.07 / Max: 3810.61. (CC) gcc options: -O3 -pthread -lz -llzma

QuantLib

QuantLib is an open-source library/framework around quantitative finance for modeling, trading and risk management scenarios. QuantLib is written in C++ with Boost and its built-in benchmark used reports the QuantLib Benchmark Index benchmark score. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Linux 5.11.3Linux 5.12 Git5001000150020002500SE +/- 38.80, N = 3SE +/- 39.65, N = 32540.62538.91. (CXX) g++ options: -O3 -march=native -rdynamic
OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.21Linux 5.11.3Linux 5.12 Git400800120016002000Min: 2465 / Avg: 2540.57 / Max: 2593.6Min: 2459.6 / Avg: 2538.9 / Max: 2579.41. (CXX) g++ options: -O3 -march=native -rdynamic

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 5.11.3Linux 5.12 Git8001600240032004000SE +/- 1.21, N = 3SE +/- 2.16, N = 33711.573709.181. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CryptoLinux 5.11.3Linux 5.12 Git6001200180024003000Min: 3709.74 / Avg: 3711.57 / Max: 3713.86Min: 3704.86 / Avg: 3709.18 / Max: 3711.471. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -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: 2MB - Disk Target: Default Test DirectoryLinux 5.12 GitLinux 5.11.330060090012001500156315621. (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 5.12 GitLinux 5.11.330060090012001500156315621. (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 5.12 GitLinux 5.11.37001400210028003500SE +/- 0.33, N = 3SE +/- 0.33, N = 3313431321. (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 5.12 GitLinux 5.11.35001000150020002500Min: 3133 / Avg: 3133.67 / Max: 3134Min: 3131 / Avg: 3131.67 / Max: 31321. (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

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantLinux 5.12 GitLinux 5.11.320K40K60K80K100KSE +/- 23.74, N = 3SE +/- 36.99, N = 390076.790133.8
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantLinux 5.12 GitLinux 5.11.316K32K48K64K80KMin: 90029.9 / Avg: 90076.67 / Max: 90107.2Min: 90073.7 / Avg: 90133.8 / Max: 90201.2

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 5.12 GitLinux 5.11.320K40K60K80K100KSE +/- 149.30, N = 3SE +/- 70.94, N = 395335.4395276.201. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Matrix MathLinux 5.12 GitLinux 5.11.317K34K51K68K85KMin: 95047.47 / Avg: 95335.43 / Max: 95547.81Min: 95195.58 / Avg: 95276.2 / Max: 95417.631. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 1080pLinux 5.11.3Linux 5.12 Git130260390520650SE +/- 2.31, N = 3SE +/- 1.55, N = 3585.81585.47MIN: 378.59 / MAX: 636.08MIN: 396.22 / MAX: 639.981. (CC) gcc options: -pthread -lm
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.2Video Input: Summer Nature 1080pLinux 5.11.3Linux 5.12 Git100200300400500Min: 581.25 / Avg: 585.81 / Max: 588.67Min: 582.73 / Avg: 585.47 / Max: 588.091. (CC) gcc options: -pthread -lm

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 5.11.3Linux 5.12 Git300K600K900K1200K1500KSE +/- 1766.48, N = 3SE +/- 4064.27, N = 31241189.11240590.9
OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000Linux 5.11.3Linux 5.12 Git200K400K600K800K1000KMin: 1237948.7 / Avg: 1241189.13 / Max: 1244028.4Min: 1234847.7 / Avg: 1240590.87 / Max: 1248444.1

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Linux 5.12 GitLinux 5.11.3400K800K1200K1600K2000KSE +/- 200.19, N = 3SE +/- 186.58, N = 318553631856223
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4Linux 5.12 GitLinux 5.11.3300K600K900K1200K1500KMin: 1855080 / Avg: 1855363.33 / Max: 1855750Min: 1855980 / Avg: 1856223.33 / Max: 1856590

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 5.11.3Linux 5.12 Git40M80M120M160M200MSE +/- 348353.95, N = 3SE +/- 369488.38, N = 3186144479.08186058640.441. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MallocLinux 5.11.3Linux 5.12 Git30M60M90M120M150MMin: 185602479.61 / Avg: 186144479.08 / Max: 186794594.7Min: 185436163.43 / Avg: 186058640.44 / Max: 186714780.051. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MMAPLinux 5.12 GitLinux 5.11.3100200300400500SE +/- 0.54, N = 3SE +/- 0.77, N = 3458.52458.471. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MMAPLinux 5.12 GitLinux 5.11.380160240320400Min: 457.48 / Avg: 458.52 / Max: 459.28Min: 456.95 / Avg: 458.47 / Max: 459.451. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatLinux 5.11.3Linux 5.12 Git20K40K60K80K100KSE +/- 43.09, N = 3SE +/- 9.22, N = 387370.287377.4
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatLinux 5.11.3Linux 5.12 Git15K30K45K60K75KMin: 87301.5 / Avg: 87370.17 / Max: 87449.6Min: 87363.4 / Avg: 87377.4 / Max: 87394.8

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 5.11.3Linux 5.12 Git30K60K90K120K150KSE +/- 2.40, N = 3SE +/- 1.83, N = 3131734.70131729.341. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Vector MathLinux 5.11.3Linux 5.12 Git20K40K60K80K100KMin: 131729.96 / Avg: 131734.7 / Max: 131737.69Min: 131726.61 / Avg: 131729.34 / Max: 131732.821. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: RdRandLinux 5.11.3Linux 5.12 Git40K80K120K160K200KSE +/- 1.81, N = 3SE +/- 9.73, N = 3184099.86184094.091. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: RdRandLinux 5.11.3Linux 5.12 Git30K60K90K120K150KMin: 184096.27 / Avg: 184099.86 / Max: 184102.01Min: 184082.04 / Avg: 184094.09 / Max: 184113.351. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

TensorFlow Lite

This is a benchmark of the TensorFlow Lite implementation. The current Linux support is limited to running on CPUs. This test profile is measuring the average inference time. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Linux 5.11.3Linux 5.12 Git400K800K1200K1600K2000KSE +/- 339.62, N = 3SE +/- 481.68, N = 316417371641773
OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2Linux 5.11.3Linux 5.12 Git300K600K900K1200K1500KMin: 1641100 / Avg: 1641736.67 / Max: 1642260Min: 1640810 / Avg: 1641773.33 / Max: 1642260

PyBench

This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesLinux 5.11.3Linux 5.12 Git2004006008001000SE +/- 1.20, N = 3912912
OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesLinux 5.11.3Linux 5.12 Git160320480640800Min: 910 / Avg: 911.67 / Max: 914

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: IO_uring - Buffered: No - Direct: Yes - Block Size: 2MB - Disk Target: Default Test DirectoryLinux 5.12 GitLinux 5.11.32004006008001000SE +/- 0.33, N = 3111711171. (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 5.12 GitLinux 5.11.32004006008001000Min: 1116 / Avg: 1116.67 / Max: 11171. (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 5.12 GitLinux 5.11.35001000150020002500SE +/- 0.58, N = 3224122411. (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 5.12 GitLinux 5.11.3400800120016002000Min: 2240 / Avg: 2241 / Max: 22421. (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 5.12 GitLinux 5.11.30.11250.2250.33750.450.5625SE +/- 0.00, N = 3SE +/- 0.00, N = 30.50.51. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncLinux 5.12 GitLinux 5.11.3246810Min: 0.5 / Avg: 0.5 / Max: 0.5Min: 0.5 / Avg: 0.5 / Max: 0.51. (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 5.11.3Linux 5.12 Git1.5M3M4.5M6M7.5MSE +/- 92839.12, N = 15SE +/- 71243.99, N = 156860046.154326056.011. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingLinux 5.11.3Linux 5.12 Git1.2M2.4M3.6M4.8M6MMin: 6229124.62 / Avg: 6860046.15 / Max: 7422431.71Min: 3860847.84 / Avg: 4326056.01 / Max: 4748767.581. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

101 Results Shown

IOR:
  256MB - Default Test Directory
  64MB - Default Test Directory
Stress-NG
IOR
PostgreSQL pgbench:
  100 - 100 - Read Write - Average Latency
  100 - 100 - Read Write
Stress-NG
FS-Mark
Compile Bench
Stress-NG
Facebook RocksDB:
  Rand Fill Sync
  Read While Writing
LevelDB
PostgreSQL pgbench
Stress-NG
Flexible IO Tester:
  Rand Write - IO_uring - No - Yes - 4KB - Default Test Directory:
    IOPS
    MB/s
  Seq Write - IO_uring - No - Yes - 4KB - Default Test Directory:
    IOPS
LevelDB
Flexible IO Tester
Stress-NG
Hackbench
FS-Mark
LevelDB:
  Overwrite
  Fill Sync
  Rand Fill
KeyDB
Zstd Compression
PostgreSQL pgbench
Stress-NG
LevelDB
FS-Mark
Compile Bench
LevelDB:
  Seq Fill
  Rand Fill
  Overwrite
Zstd Compression
Compile Bench
Ngspice
Facebook RocksDB
Zstd Compression
Facebook RocksDB
dav1d
LevelDB
Stress-NG
PostgreSQL pgbench
Stress-NG
PostgreSQL pgbench:
  100 - 100 - Read Only
  100 - 50 - Read Write
Timed Wasmer Compilation
PostgreSQL pgbench
Zstd Compression
ctx_clock
OpenFOAM
Flexible IO Tester:
  Seq Read - IO_uring - No - Yes - 4KB - Default Test Directory:
    MB/s
    IOPS
Ngspice
InfluxDB
Zstd Compression
Flexible IO Tester
Timed Linux Kernel Compilation
Stress-NG
LevelDB
Stress-NG:
  CPU Stress
  Glibc Qsort Data Sorting
TensorFlow Lite
Facebook RocksDB
Flexible IO Tester:
  Seq Write - IO_uring - No - Yes - 2MB - Default Test Directory
  Rand Read - IO_uring - No - Yes - 4KB - Default Test Directory
  Seq Write - IO_uring - No - Yes - 2MB - Default Test Directory
InfluxDB
Timed Erlang/OTP Compilation
GROMACS
Stress-NG
PHPBench
TensorFlow Lite
NAMD
Stress-NG
Flexible IO Tester
Zstd Compression
QuantLib
Stress-NG
Flexible IO Tester:
  Seq Read - IO_uring - No - Yes - 2MB - Default Test Directory
  Rand Read - IO_uring - No - Yes - 2MB - Default Test Directory
  Rand Read - IO_uring - No - Yes - 2MB - Default Test Directory
TensorFlow Lite
Stress-NG
dav1d
InfluxDB
TensorFlow Lite
Stress-NG:
  Malloc
  MMAP
TensorFlow Lite
Stress-NG:
  Vector Math
  RdRand
TensorFlow Lite
PyBench
Flexible IO Tester:
  Rand Write - IO_uring - No - Yes - 2MB - Default Test Directory:
    IOPS
    MB/s
LevelDB
Stress-NG

Linux 5.11.3

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: llvmpipe, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211

OS: Ubuntu 20.04, Kernel: 5.11.3-051103-generic (x86_64), Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 3.3 Mesa 20.0.8 (LLVM 10.0.0 256 bits), Compiler: GCC 9.3.0 + CUDA 11.1, File-System: ext4, Screen Resolution: 1024x768

Kernel Notes: Transparent Huge Pages: madvise
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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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: intel_cpufreq ondemand - CPU Microcode: 0x5003003
Python Notes: Python 2.7.18 + Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 5 March 2021 06:54 by user pts.

Linux 5.12 Git

Processor: Intel Core i9-10980XE @ 4.80GHz (18 Cores / 36 Threads), Motherboard: ASRock X299 Steel Legend (P1.30 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 32GB, Disk: Samsung SSD 970 PRO 512GB, Graphics: llvmpipe, Audio: Realtek ALC1220, Network: Intel I219-V + Intel I211

OS: Ubuntu 20.04, Kernel: 5.12.0-051200rc1daily20210305-generic (x86_64) 20210304, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.9, OpenGL: 3.3 Mesa 20.0.8 (LLVM 10.0.0 256 bits), Compiler: GCC 9.3.0 + CUDA 11.1, File-System: ext4, Screen Resolution: 1024x768

Kernel Notes: Transparent Huge Pages: madvise
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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-9-HskZEa/gcc-9-9.3.0/debian/tmp-nvptx/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: intel_cpufreq ondemand - CPU Microcode: 0x5003003
Python Notes: Python 2.7.18 + Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + 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 Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 5 March 2021 12:19 by user pts.