Haswell 2021

Intel Xeon E5-2687W v3 testing with a MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS) and NVIDIA GeForce GTX 770 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 2101287-HA-HASWELL2015
Jump To Table - Results

View

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

Limit displaying results to tests within:

Audio Encoding 2 Tests
AV1 2 Tests
C++ Boost Tests 3 Tests
Timed Code Compilation 3 Tests
C/C++ Compiler Tests 3 Tests
CPU Massive 10 Tests
Creator Workloads 9 Tests
Cryptography 4 Tests
Encoding 4 Tests
Finance 2 Tests
Fortran Tests 3 Tests
Game Development 2 Tests
HPC - High Performance Computing 15 Tests
Machine Learning 5 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 4 Tests
Multi-Core 10 Tests
NVIDIA GPU Compute 2 Tests
OpenMPI Tests 9 Tests
Programmer / Developer System Benchmarks 5 Tests
Python Tests 2 Tests
Scientific Computing 8 Tests
Server CPU Tests 6 Tests
Single-Threaded 3 Tests
Video Encoding 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
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
1
January 26 2021
  8 Hours, 35 Minutes
2
January 27 2021
  9 Hours, 32 Minutes
3
January 27 2021
  8 Hours, 55 Minutes
Invert Hiding All Results Option
  9 Hours, 1 Minute

Only show results where is faster than
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):

1

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
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x44
Python Notes: Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

2

3

Processor: Intel Xeon E5-2687W v3 @ 3.50GHz (10 Cores / 20 Threads), Motherboard: MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS), Chipset: Intel Xeon E7 v3/Xeon, Memory: 32GB, Disk: 80GB INTEL SSDSCKGW08, Graphics: NVIDIA GeForce GTX 770, Audio: Realtek ALC892, Monitor: LG Ultra HD, Network: Intel I218-V

OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc7daily20200928-generic (x86_64) 20200927, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: modesetting 1.20.8, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 3840x2160

Haswell 2021ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen Resolution123Intel Xeon E5-2687W v3 @ 3.50GHz (10 Cores / 20 Threads)MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS)Intel Xeon E7 v3/Xeon32GB80GB INTEL SSDSCKGW08NVIDIA GeForce GTX 770Realtek ALC892LG Ultra HDIntel I218-VUbuntu 20.045.9.0-050900rc7daily20200928-generic (x86_64) 20200927GNOME Shell 3.36.4X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext43840x2160OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --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 Processor Details- Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x44Python Details- Python 3.8.5Security Details- itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

123Result OverviewPhoronix Test Suite100%105%110%115%CLOMPRedisLAMMPS Molecular Dynamics SimulatorNAS Parallel BenchmarksCP2K Molecular DynamicsCloverLeafOpus Codec EncodingKripkeQMCPACKCpuminer-OptGnuPGTimed Eigen Compilationperf-benchUnpacking FirefoxAlgebraic Multi-Grid BenchmarkLULESHBuild2ASKAPNCNNOpenFOAMWavPack Audio EncodingTNNWebP2 Image EncodeGcrypt Librarydav1dCryptsetupGoogle SynthMarkoneDNNONNX RuntimeTimed Godot Game Engine CompilationlzbenchMobile Neural NetworkQuantLibrav1eEtcpakFinanceBench

Haswell 2021redis: GETnpb: EP.Cncnn: CPU - yolov4-tinycpuminer-opt: Skeincoincloverleaf: Lagrangian-Eulerian Hydrodynamicsfinancebench: Repo OpenMPonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUcpuminer-opt: Blake-2 Sncnn: CPU - resnet18financebench: Bonds OpenMPcp2k: Fayalite-FIST Datancnn: CPU - googlenetredis: SETaskap: tConvolve OpenMP - Degriddingcpuminer-opt: Myriad-Groestllzbench: Zstd 8 - Decompressionnpb: EP.Dencode-opus: WAV To Opus Encodemnn: MobileNetV2_224webp2: Defaultkripke: qmcpack: simple-H2Oncnn: CPU - blazefacencnn: CPU-v2-v2 - mobilenet-v2ncnn: CPU - regnety_400mmnn: mobilenet-v1-1.0dav1d: Summer Nature 1080pgnupg: 2.7GB Sample File Encryptionbuild-eigen: Time To Compiledav1d: Summer Nature 4Kcpuminer-opt: Magiredis: SADDmnn: resnet-v2-50perf-bench: Sched Pipeunpack-firefox: firefox-84.0.source.tar.xzncnn: CPU - mobilenetncnn: CPU - resnet50cpuminer-opt: Garlicoincryptsetup: AES-XTS 256b Decryptionamg: onnx: bertsquad-10 - OpenMP CPUncnn: CPU - mnasnetcryptsetup: AES-XTS 256b Encryptionlulesh: cpuminer-opt: Ringcoinlzbench: Brotli 2 - Decompressionbuild2: Time To Compileaskap: tConvolve MT - Degriddingredis: LPUSHncnn: CPU - efficientnet-b0lzbench: Brotli 0 - Decompressionrav1e: 10onednn: Recurrent Neural Network Inference - u8s8f32 - CPUtnn: CPU - SqueezeNet v1.1cpuminer-opt: LBC, LBRY Creditsncnn: CPU - shufflenet-v2onednn: Recurrent Neural Network Inference - f32 - CPUcpuminer-opt: x25xonednn: IP Shapes 3D - f32 - CPUcryptsetup: Twofish-XTS 256b Encryptioncryptsetup: PBKDF2-whirlpooltnn: CPU - MobileNet v2onnx: shufflenet-v2-10 - OpenMP CPUrav1e: 6onednn: Matrix Multiply Batch Shapes Transformer - u8s8f32 - CPUcryptsetup: Serpent-XTS 256b Encryptiononednn: IP Shapes 3D - u8s8f32 - CPUencode-wavpack: WAV To WavPackgcrypt: dav1d: Chimera 1080p 10-bitetcpak: ETC1cryptsetup: Serpent-XTS 256b Decryptionetcpak: ETC2ncnn: CPU-v3-v3 - mobilenet-v3ncnn: CPU - vgg16rav1e: 1etcpak: DXT1cryptsetup: AES-XTS 512b Decryptionncnn: CPU - alexnetaskap: tConvolve MT - Griddingcryptsetup: Twofish-XTS 512b Decryptioncryptsetup: Twofish-XTS 256b Decryptiononnx: yolov4 - OpenMP CPUlzbench: Zstd 1 - Decompressioncryptsetup: AES-XTS 512b Encryptiononednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUlzbench: Brotli 0 - Compressiononednn: Matrix Multiply Batch Shapes Transformer - f32 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUdav1d: Chimera 1080pmnn: SqueezeNetV1.0lzbench: Zstd 1 - Compressionrav1e: 5webp2: Quality 95, Compression Effort 7openfoam: Motorbike 30Mlzbench: Crush 0 - Decompressionsynthmark: VoiceMark_100onednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUwebp2: Quality 75, Compression Effort 7cryptsetup: Serpent-XTS 512b Decryptiononednn: Convolution Batch Shapes Auto - u8s8f32 - CPUcryptsetup: Serpent-XTS 512b Encryptionmnn: inception-v3etcpak: ETC1 + Ditheringaskap: Hogbom Clean OpenMPbuild-godot: Time To Compilencnn: CPU - squeezenet_ssdwebp2: Quality 100, Lossless Compressiononednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - f32 - CPUonednn: IP Shapes 1D - f32 - CPUonednn: Recurrent Neural Network Training - u8s8f32 - CPUquantlib: lzbench: Libdeflate 1 - Decompressioncryptsetup: PBKDF2-sha512onnx: super-resolution-10 - OpenMP CPUonednn: Recurrent Neural Network Training - f32 - CPUwebp2: Quality 100, Compression Effort 5onednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUcryptsetup: Twofish-XTS 512b Encryptiononnx: fcn-resnet101-11 - OpenMP CPUlzbench: Libdeflate 1 - Compressionlzbench: Brotli 2 - Compressionlzbench: Crush 0 - Compressionlzbench: Zstd 8 - Compressionlzbench: XZ 0 - Decompressionlzbench: XZ 0 - Compressionredis: LPOPaskap: tConvolve OpenMP - Griddingaskap: tConvolve MPI - Griddingaskap: tConvolve MPI - Degriddingcpuminer-opt: Triple SHA-256, Onecoincpuminer-opt: Quad SHA-256, Pyritecpuminer-opt: Deepcoinlammps: Rhodopsin Proteinclomp: Static OMP Speedup1231887226.871050.1731.4334847136.8968017.6536466.6601122050815.20115661.1296871543.68316.601415197.082664.351036314231027.8810.6865.0615.6774610442051.1532.766.2224.086.196382.2181.081109.859142.04203.981602736.8854.5117375524.03820.0329.061736.891697.23028911335455.521702.34654.46641579.48581159.6871756.701247740.008.285032.3752208.25329.721233307.052209.39219.587.41691344.4530142339.21297361.0653.00389549.02.4899417.395273.55168.44235.540532.9140.7315.4151.780.2771083.8611389.612.081285.02346.4346.832813901400.24051.993563.368502.83074459.118.4474010.810545.956219.05431551.1818.393255.76536298.918533.712.7075550.755.865226.752239.235186.69726.55939.7605.0756413.78544.074764045.791689.6995132899340074048.9514.4822211.56345.659182148766596342010121.551778.102071.461699.8561247476537487.984.41513.21775802.501002.0131.9634338140.6469828.3619796.5193722193815.54114077.5651041576.95216.841408307.622716.91056714131008.9110.6175.1465.6644652985351.9032.796.2223.936.112380.8680.221111.128140.67204.181588662.7553.9667369523.89120.2128.861750.881702.63036123675415.551708.54644.46011582.49585160.6741753.611242294.798.305062.3612211.70328.625234077.072214.44220.737.39027345.4532814337.51197751.0702.99080550.32.4793017.322272.41168.62235.968531.5140.3155.4151.970.2771080.5461393.012.121284.71346.8345.832813891401.74040.593563.359472.83861459.328.4394010.810545.663219.10432550.1428.401245.77823298.253533.312.6938551.255.917227.136239.425186.84426.57938.9165.0771413.79854.074514044.151691.5994133011840084050.5414.4782211.32345.759182148766596341271416.381881.941984.731589.7560350486227013.874.26111.01765664.371047.6732.7335707140.5769374.8691416.6841122607315.19116695.3782551546.67816.951436898.962716.91051013961019.3710.7965.1005.7564684042751.9132.756.3123.756.151385.9980.127111.065140.47206.241605411.5854.1247443024.12720.0629.111738.451710.63051336335425.561714.44621.63051590.60582160.0931745.941249824.008.335052.3752221.13327.826234637.032221.93220.827.43079346.2532634338.74197291.0682.99407551.42.4812517.326273.01368.34236.454533.5140.2105.4351.850.2761084.3291394.312.111280.90345.7346.932713861404.24048.113573.368832.83373458.068.4244000.808544.662219.57431549.9358.382265.76890298.540532.612.7193551.755.960227.056239.617186.55726.59940.2315.0700513.78044.079804040.611691.6995132899340054051.4214.4752210.61345.659182148766596341758555.231962.741979.001597.8462923483417312.244.46313.0OpenBenchmarking.org

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GET123400K800K1200K1600K2000KSE +/- 6745.04, N = 3SE +/- 25400.58, N = 3SE +/- 20496.02, N = 31887226.871775802.501765664.371. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: GET123300K600K900K1200K1500KMin: 1874771.62 / Avg: 1887226.87 / Max: 1897941.88Min: 1725035.75 / Avg: 1775802.5 / Max: 1802805.12Min: 1732040.12 / Avg: 1765664.37 / Max: 1802782.121. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.C1232004006008001000SE +/- 16.48, N = 3SE +/- 11.00, N = 15SE +/- 4.68, N = 31050.171002.011047.671. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.C1232004006008001000Min: 1020.07 / Avg: 1050.17 / Max: 1076.86Min: 946.77 / Avg: 1002.01 / Max: 1065.22Min: 1039.72 / Avg: 1047.67 / Max: 1055.931. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tiny123816243240SE +/- 0.35, N = 3SE +/- 0.60, N = 3SE +/- 0.34, N = 331.4331.9632.73MIN: 30.14 / MAX: 34.58MIN: 30.06 / MAX: 35.42MIN: 31.73 / MAX: 35.091. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: yolov4-tiny123714212835Min: 30.74 / Avg: 31.43 / Max: 31.8Min: 30.82 / Avg: 31.96 / Max: 32.83Min: 32.13 / Avg: 32.73 / Max: 33.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Skeincoin1238K16K24K32K40KSE +/- 291.68, N = 15SE +/- 325.79, N = 15SE +/- 140.75, N = 33484734338357071. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Skeincoin1236K12K18K24K30KMin: 32900 / Avg: 34847.33 / Max: 36170Min: 32250 / Avg: 34338 / Max: 36610Min: 35430 / Avg: 35706.67 / Max: 358901. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm.in input file (Problem 5). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian Hydrodynamics123306090120150SE +/- 0.22, N = 3SE +/- 0.54, N = 3SE +/- 0.33, N = 3136.89140.64140.571. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian Hydrodynamics123306090120150Min: 136.64 / Avg: 136.89 / Max: 137.34Min: 139.64 / Avg: 140.64 / Max: 141.51Min: 140.11 / Avg: 140.57 / Max: 141.221. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMP12315K30K45K60K75KSE +/- 89.04, N = 3SE +/- 1002.67, N = 3SE +/- 922.31, N = 468017.6569828.3669374.871. (CXX) g++ options: -O3 -march=native -fopenmp
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Repo OpenMP12312K24K36K48K60KMin: 67849.05 / Avg: 68017.65 / Max: 68151.59Min: 67859.35 / Avg: 69828.36 / Max: 71141.9Min: 67864.23 / Avg: 69374.87 / Max: 71784.761. (CXX) g++ options: -O3 -march=native -fopenmp

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU123246810SE +/- 0.08426, N = 4SE +/- 0.07423, N = 6SE +/- 0.09722, N = 36.660116.519376.68411MIN: 6.33MIN: 6.3MIN: 6.421. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPU1233691215Min: 6.41 / Avg: 6.66 / Max: 6.76Min: 6.36 / Avg: 6.52 / Max: 6.75Min: 6.5 / Avg: 6.68 / Max: 6.821. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Blake-2 S12350K100K150K200K250KSE +/- 3265.93, N = 4SE +/- 2626.85, N = 15SE +/- 3030.09, N = 42205082219382260731. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Blake-2 S12340K80K120K160K200KMin: 213000 / Avg: 220507.5 / Max: 226890Min: 200870 / Avg: 221938 / Max: 242440Min: 217470 / Avg: 226072.5 / Max: 2306601. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet1812348121620SE +/- 0.04, N = 3SE +/- 0.15, N = 3SE +/- 0.06, N = 315.2015.5415.19MIN: 15.01 / MAX: 15.36MIN: 15 / MAX: 110.46MIN: 14.96 / MAX: 16.481. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet1812348121620Min: 15.11 / Avg: 15.2 / Max: 15.25Min: 15.27 / Avg: 15.54 / Max: 15.78Min: 15.13 / Avg: 15.19 / Max: 15.31. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

FinanceBench

FinanceBench is a collection of financial program benchmarks with support for benchmarking on the GPU via OpenCL and CPU benchmarking with OpenMP. The FinanceBench test cases are focused on Black-Sholes-Merton Process with Analytic European Option engine, QMC (Sobol) Monte-Carlo method (Equity Option Example), Bonds Fixed-rate bond with flat forward curve, and Repo Securities repurchase agreement. FinanceBench was originally written by the Cavazos Lab at University of Delaware. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMP12320K40K60K80K100KSE +/- 1477.66, N = 5SE +/- 20.05, N = 3SE +/- 1452.56, N = 12115661.13114077.57116695.381. (CXX) g++ options: -O3 -march=native -fopenmp
OpenBenchmarking.orgms, Fewer Is BetterFinanceBench 2016-07-25Benchmark: Bonds OpenMP12320K40K60K80K100KMin: 114070.01 / Avg: 115661.13 / Max: 121569.52Min: 114041.66 / Avg: 114077.57 / Max: 114110.98Min: 113998.14 / Avg: 116695.38 / Max: 130678.231. (CXX) g++ options: -O3 -march=native -fopenmp

CP2K Molecular Dynamics

CP2K is an open-source molecular dynamics software package focused on quantum chemistry and solid-state physics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 8.1Fayalite-FIST Data123300600900120015001543.681576.951546.68

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenet12348121620SE +/- 0.14, N = 3SE +/- 0.20, N = 3SE +/- 0.31, N = 316.6016.8416.95MIN: 16.2 / MAX: 74.59MIN: 16.12 / MAX: 18.21MIN: 16.16 / MAX: 17.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: googlenet12348121620Min: 16.45 / Avg: 16.6 / Max: 16.88Min: 16.43 / Avg: 16.84 / Max: 17.05Min: 16.63 / Avg: 16.95 / Max: 17.561. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SET123300K600K900K1200K1500KSE +/- 14514.92, N = 8SE +/- 19369.44, N = 15SE +/- 9026.88, N = 31415197.081408307.621436898.961. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SET123200K400K600K800K1000KMin: 1329450.75 / Avg: 1415197.08 / Max: 1456484.62Min: 1188087 / Avg: 1408307.62 / Max: 1463486Min: 1425724.62 / Avg: 1436898.96 / Max: 1454766.381. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - Degridding1236001200180024003000SE +/- 1.79, N = 15SE +/- 0.00, N = 15SE +/- 0.00, N = 32664.352716.902716.901. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - Degridding1235001000150020002500Min: 2662.56 / Avg: 2664.35 / Max: 2689.45Min: 2716.9 / Avg: 2716.9 / Max: 2716.9Min: 2716.9 / Avg: 2716.9 / Max: 2716.91. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Myriad-Groestl1232K4K6K8K10KSE +/- 6.67, N = 3SE +/- 161.69, N = 3SE +/- 125.83, N = 31036310567105101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Myriad-Groestl1232K4K6K8K10KMin: 10350 / Avg: 10363.33 / Max: 10370Min: 10400 / Avg: 10566.67 / Max: 10890Min: 10360 / Avg: 10510 / Max: 107601. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: Decompression12330060090012001500SE +/- 4.06, N = 3SE +/- 7.88, N = 3SE +/- 21.06, N = 31423141313961. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: Decompression1232004006008001000Min: 1416 / Avg: 1423.33 / Max: 1430Min: 1398 / Avg: 1412.67 / Max: 1425Min: 1355 / Avg: 1396.33 / Max: 14241. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. This test profile offers selecting the different NPB tests/problems and varying problem sizes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.D1232004006008001000SE +/- 14.12, N = 4SE +/- 9.01, N = 12SE +/- 17.23, N = 31027.881008.911019.371. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.D1232004006008001000Min: 994 / Avg: 1027.88 / Max: 1059.83Min: 934.42 / Avg: 1008.91 / Max: 1037.42Min: 994.34 / Avg: 1019.37 / Max: 1052.391. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 4.0.3

Opus Codec Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encode1233691215SE +/- 0.03, N = 5SE +/- 0.04, N = 5SE +/- 0.03, N = 510.6910.6210.801. (CXX) g++ options: -fvisibility=hidden -logg -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus Encode1233691215Min: 10.62 / Avg: 10.69 / Max: 10.75Min: 10.48 / Avg: 10.62 / Max: 10.74Min: 10.68 / Avg: 10.8 / Max: 10.851. (CXX) g++ options: -fvisibility=hidden -logg -lm

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: MobileNetV2_2241231.15792.31583.47374.63165.7895SE +/- 0.016, N = 3SE +/- 0.008, N = 3SE +/- 0.037, N = 35.0615.1465.100MIN: 4.98 / MAX: 5.99MIN: 5.03 / MAX: 6.03MIN: 4.42 / MAX: 11.391. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: MobileNetV2_224123246810Min: 5.03 / Avg: 5.06 / Max: 5.08Min: 5.13 / Avg: 5.15 / Max: 5.16Min: 5.04 / Avg: 5.1 / Max: 5.171. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Default1231.29512.59023.88535.18046.4755SE +/- 0.037, N = 3SE +/- 0.018, N = 3SE +/- 0.043, N = 35.6775.6645.7561. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg
OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Default123246810Min: 5.61 / Avg: 5.68 / Max: 5.73Min: 5.63 / Avg: 5.66 / Max: 5.69Min: 5.67 / Avg: 5.76 / Max: 5.811. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg

Kripke

Kripke is a simple, scalable, 3D Sn deterministic particle transport code. Its primary purpose is to research how data layout, programming paradigms and architectures effect the implementation and performance of Sn transport. Kripke is developed by LLNL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.412310M20M30M40M50MSE +/- 200225.59, N = 3SE +/- 199447.85, N = 3SE +/- 186245.67, N = 34610442046529853468404271. (CXX) g++ options: -O3 -fopenmp
OpenBenchmarking.orgThroughput FoM, More Is BetterKripke 1.2.41238M16M24M32M40MMin: 45711850 / Avg: 46104420 / Max: 46369170Min: 46182630 / Avg: 46529853.33 / Max: 46873510Min: 46608990 / Avg: 46840426.67 / Max: 472089101. (CXX) g++ options: -O3 -fopenmp

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. QMCPACK is an open-source production level many-body ab initio Quantum Monte Carlo code for computing the electronic structure of atoms, molecules, and solids. QMCPACK is supported by the U.S. Department of Energy. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.10Input: simple-H2O1231224364860SE +/- 0.64, N = 5SE +/- 0.70, N = 5SE +/- 0.74, N = 351.1551.9051.911. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm
OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.10Input: simple-H2O1231020304050Min: 49.97 / Avg: 51.15 / Max: 52.75Min: 49.81 / Avg: 51.9 / Max: 54.17Min: 50.46 / Avg: 51.91 / Max: 52.911. (CXX) g++ options: -fopenmp -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -march=native -O3 -fomit-frame-pointer -ffast-math -pthread -lm

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazeface1230.62781.25561.88342.51123.139SE +/- 0.01, N = 3SE +/- 0.04, N = 3SE +/- 0.02, N = 32.762.792.75MIN: 2.7 / MAX: 3.13MIN: 2.69 / MAX: 3.4MIN: 2.69 / MAX: 3.111. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: blazeface123246810Min: 2.75 / Avg: 2.76 / Max: 2.77Min: 2.71 / Avg: 2.79 / Max: 2.83Min: 2.72 / Avg: 2.75 / Max: 2.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v2123246810SE +/- 0.01, N = 3SE +/- 0.02, N = 3SE +/- 0.10, N = 36.226.226.31MIN: 6.14 / MAX: 6.38MIN: 6.12 / MAX: 6.83MIN: 6.11 / MAX: 64.271. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v2-v2 - Model: mobilenet-v21233691215Min: 6.2 / Avg: 6.22 / Max: 6.24Min: 6.19 / Avg: 6.22 / Max: 6.24Min: 6.19 / Avg: 6.31 / Max: 6.511. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400m123612182430SE +/- 0.03, N = 3SE +/- 0.13, N = 3SE +/- 0.06, N = 324.0823.9323.75MIN: 23.89 / MAX: 25.07MIN: 23.59 / MAX: 24.31MIN: 23.54 / MAX: 24.431. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: regnety_400m123612182430Min: 24.05 / Avg: 24.08 / Max: 24.14Min: 23.68 / Avg: 23.93 / Max: 24.08Min: 23.64 / Avg: 23.75 / Max: 23.831. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: mobilenet-v1-1.0123246810SE +/- 0.006, N = 3SE +/- 0.007, N = 3SE +/- 0.018, N = 36.1966.1126.151MIN: 4.66 / MAX: 19.35MIN: 6.03 / MAX: 12.61MIN: 6.05 / MAX: 6.91. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: mobilenet-v1-1.0123246810Min: 6.19 / Avg: 6.2 / Max: 6.21Min: 6.11 / Avg: 6.11 / Max: 6.13Min: 6.12 / Avg: 6.15 / Max: 6.181. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

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 0.8.1Video Input: Summer Nature 1080p12380160240320400SE +/- 3.01, N = 3SE +/- 3.16, N = 3SE +/- 1.19, N = 3382.21380.86385.99MIN: 305.4 / MAX: 420.06MIN: 302.45 / MAX: 418.3MIN: 312.93 / MAX: 423.31. (CC) gcc options: -pthread
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Summer Nature 1080p12370140210280350Min: 376.47 / Avg: 382.21 / Max: 386.67Min: 374.55 / Avg: 380.86 / Max: 384.19Min: 384.3 / Avg: 385.99 / Max: 388.291. (CC) gcc options: -pthread

GnuPG

This test times how long it takes to encrypt a sample file using GnuPG. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGnuPG 2.2.272.7GB Sample File Encryption12320406080100SE +/- 1.03, N = 3SE +/- 0.16, N = 3SE +/- 0.07, N = 381.0880.2280.131. (CC) gcc options: -O2
OpenBenchmarking.orgSeconds, Fewer Is BetterGnuPG 2.2.272.7GB Sample File Encryption1231530456075Min: 79.95 / Avg: 81.08 / Max: 83.14Min: 80 / Avg: 80.22 / Max: 80.53Min: 80.04 / Avg: 80.13 / Max: 80.261. (CC) gcc options: -O2

Timed Eigen Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To Compile12320406080100SE +/- 0.16, N = 3SE +/- 0.09, N = 3SE +/- 0.15, N = 3109.86111.13111.07
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Eigen Compilation 3.3.9Time To Compile12320406080100Min: 109.59 / Avg: 109.86 / Max: 110.16Min: 110.99 / Avg: 111.13 / Max: 111.29Min: 110.78 / Avg: 111.07 / Max: 111.23

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 0.8.1Video Input: Summer Nature 4K123306090120150SE +/- 0.27, N = 3SE +/- 0.53, N = 3SE +/- 0.14, N = 3142.04140.67140.47MIN: 130.3 / MAX: 163.29MIN: 125.57 / MAX: 161.64MIN: 130.58 / MAX: 160.851. (CC) gcc options: -pthread
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Summer Nature 4K123306090120150Min: 141.76 / Avg: 142.04 / Max: 142.57Min: 139.67 / Avg: 140.67 / Max: 141.48Min: 140.2 / Avg: 140.47 / Max: 140.691. (CC) gcc options: -pthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Magi12350100150200250SE +/- 0.12, N = 3SE +/- 0.17, N = 3SE +/- 2.29, N = 14203.98204.18206.241. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Magi1234080120160200Min: 203.75 / Avg: 203.98 / Max: 204.13Min: 203.86 / Avg: 204.18 / Max: 204.41Min: 202.87 / Avg: 206.24 / Max: 235.961. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SADD123300K600K900K1200K1500KSE +/- 21115.79, N = 4SE +/- 10942.24, N = 3SE +/- 7906.68, N = 31602736.881588662.751605411.581. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: SADD123300K600K900K1200K1500KMin: 1539413.88 / Avg: 1602736.88 / Max: 1625223.5Min: 1569371 / Avg: 1588662.75 / Max: 1607256.75Min: 1594398 / Avg: 1605411.58 / Max: 1620745.51. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: resnet-v2-501231224364860SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.10, N = 354.5153.9754.12MIN: 53.53 / MAX: 130.53MIN: 53.45 / MAX: 128.98MIN: 53.84 / MAX: 125.641. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: resnet-v2-501231122334455Min: 54.41 / Avg: 54.51 / Max: 54.64Min: 53.9 / Avg: 53.97 / Max: 54.08Min: 53.98 / Avg: 54.12 / Max: 54.331. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

perf-bench

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

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched Pipe12316K32K48K64K80KSE +/- 327.32, N = 3SE +/- 929.56, N = 5SE +/- 629.67, N = 37375573695744301. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched Pipe12313K26K39K52K65KMin: 73261 / Avg: 73755 / Max: 74374Min: 71790 / Avg: 73695.4 / Max: 76774Min: 73248 / Avg: 74430.33 / Max: 753971. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -Xlinker -lpthread -lrt -lm -ldl -lelf -lcrypto -lz -llzma -lnuma

Unpacking Firefox

This simple test profile measures how long it takes to extract the .tar.xz source package of the Mozilla Firefox Web Browser. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterUnpacking Firefox 84.0Extracting: firefox-84.0.source.tar.xz123612182430SE +/- 0.17, N = 4SE +/- 0.07, N = 4SE +/- 0.10, N = 424.0423.8924.13
OpenBenchmarking.orgSeconds, Fewer Is BetterUnpacking Firefox 84.0Extracting: firefox-84.0.source.tar.xz123612182430Min: 23.7 / Avg: 24.04 / Max: 24.47Min: 23.73 / Avg: 23.89 / Max: 24.06Min: 23.99 / Avg: 24.13 / Max: 24.44

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenet123510152025SE +/- 0.02, N = 3SE +/- 0.14, N = 3SE +/- 0.01, N = 320.0320.2120.06MIN: 19.87 / MAX: 21.34MIN: 19.81 / MAX: 21.32MIN: 19.95 / MAX: 21.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mobilenet123510152025Min: 20 / Avg: 20.03 / Max: 20.06Min: 20 / Avg: 20.21 / Max: 20.47Min: 20.04 / Avg: 20.06 / Max: 20.081. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50123714212835SE +/- 0.25, N = 3SE +/- 0.10, N = 3SE +/- 0.09, N = 329.0628.8629.11MIN: 28.13 / MAX: 134.5MIN: 28.14 / MAX: 29.56MIN: 28.24 / MAX: 30.151. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: resnet50123612182430Min: 28.8 / Avg: 29.06 / Max: 29.55Min: 28.66 / Avg: 28.86 / Max: 29.01Min: 28.93 / Avg: 29.11 / Max: 29.251. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Garlicoin123400800120016002000SE +/- 7.48, N = 3SE +/- 6.73, N = 3SE +/- 4.72, N = 31736.891750.881738.451. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Garlicoin12330060090012001500Min: 1722 / Avg: 1736.89 / Max: 1745.58Min: 1738.98 / Avg: 1750.88 / Max: 1762.29Min: 1730.24 / Avg: 1738.45 / Max: 1746.591. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Decryption123400800120016002000SE +/- 8.30, N = 3SE +/- 1.84, N = 3SE +/- 1.95, N = 31697.21702.61710.6
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Decryption12330060090012001500Min: 1680.7 / Avg: 1697.17 / Max: 1707.2Min: 1699.1 / Avg: 1702.63 / Max: 1705.3Min: 1708 / Avg: 1710.57 / Max: 1714.4

Algebraic Multi-Grid Benchmark

AMG is a parallel algebraic multigrid solver for linear systems arising from problems on unstructured grids. The driver provided with AMG builds linear systems for various 3-dimensional problems. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.212370M140M210M280M350MSE +/- 2829162.69, N = 3SE +/- 2849594.82, N = 3SE +/- 980611.02, N = 33028911333036123673051336331. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi
OpenBenchmarking.orgFigure Of Merit, More Is BetterAlgebraic Multi-Grid Benchmark 1.212350M100M150M200M250MMin: 297253600 / Avg: 302891133.33 / Max: 306129600Min: 297934400 / Avg: 303612366.67 / Max: 306876900Min: 303233600 / Avg: 305133633.33 / Max: 3065046001. (CC) gcc options: -lparcsr_ls -lparcsr_mv -lseq_mv -lIJ_mv -lkrylov -lHYPRE_utilities -lm -fopenmp -pthread -lmpi

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: bertsquad-10 - Device: OpenMP CPU123120240360480600SE +/- 0.87, N = 3SE +/- 1.36, N = 3SE +/- 0.83, N = 35455415421. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: bertsquad-10 - Device: OpenMP CPU123100200300400500Min: 543.5 / Avg: 545 / Max: 546.5Min: 539.5 / Avg: 541.33 / Max: 544Min: 541 / Avg: 541.83 / Max: 543.51. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnet1231.2512.5023.7535.0046.255SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 35.525.555.56MIN: 5.41 / MAX: 5.78MIN: 5.4 / MAX: 5.65MIN: 5.41 / MAX: 5.791. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: mnasnet123246810Min: 5.5 / Avg: 5.52 / Max: 5.53Min: 5.54 / Avg: 5.55 / Max: 5.56Min: 5.48 / Avg: 5.56 / Max: 5.691. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Encryption123400800120016002000SE +/- 6.83, N = 3SE +/- 3.81, N = 3SE +/- 2.43, N = 31702.31708.51714.4
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Encryption12330060090012001500Min: 1689 / Avg: 1702.27 / Max: 1711.7Min: 1701.2 / Avg: 1708.53 / Max: 1714Min: 1709.8 / Avg: 1714.37 / Max: 1718.1

LULESH

LULESH is the Livermore Unstructured Lagrangian Explicit Shock Hydrodynamics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.312310002000300040005000SE +/- 37.60, N = 3SE +/- 60.83, N = 3SE +/- 60.75, N = 34654.474644.464621.631. (CXX) g++ options: -O3 -fopenmp -lm -pthread -lmpi_cxx -lmpi
OpenBenchmarking.orgz/s, More Is BetterLULESH 2.0.31238001600240032004000Min: 4579.27 / Avg: 4654.47 / Max: 4692.72Min: 4522.93 / Avg: 4644.46 / Max: 4710.15Min: 4500.2 / Avg: 4621.63 / Max: 4685.751. (CXX) g++ options: -O3 -fopenmp -lm -pthread -lmpi_cxx -lmpi

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Ringcoin12330060090012001500SE +/- 3.43, N = 3SE +/- 10.01, N = 3SE +/- 11.74, N = 141579.481582.491590.601. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Ringcoin12330060090012001500Min: 1573.02 / Avg: 1579.48 / Max: 1584.72Min: 1568.63 / Avg: 1582.49 / Max: 1601.94Min: 1564.61 / Avg: 1590.6 / Max: 1741.411. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: Decompression123130260390520650SE +/- 5.17, N = 3SE +/- 0.33, N = 3SE +/- 2.73, N = 35815855821. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: Decompression123100200300400500Min: 571 / Avg: 581.33 / Max: 587Min: 585 / Avg: 585.33 / Max: 586Min: 577 / Avg: 582.33 / Max: 5861. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Build2

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

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compile1234080120160200SE +/- 0.55, N = 3SE +/- 0.85, N = 3SE +/- 0.91, N = 3159.69160.67160.09
OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.13Time To Compile123306090120150Min: 158.78 / Avg: 159.69 / Max: 160.66Min: 159.34 / Avg: 160.67 / Max: 162.26Min: 158.31 / Avg: 160.09 / Max: 161.29

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - Degridding123400800120016002000SE +/- 0.84, N = 3SE +/- 0.77, N = 3SE +/- 0.99, N = 31756.701753.611745.941. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - Degridding12330060090012001500Min: 1755.15 / Avg: 1756.7 / Max: 1758.05Min: 1752.84 / Avg: 1753.61 / Max: 1755.15Min: 1744.23 / Avg: 1745.94 / Max: 1747.661. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPUSH123300K600K900K1200K1500KSE +/- 3115.55, N = 3SE +/- 9099.06, N = 3SE +/- 2768.29, N = 31247740.001242294.791249824.001. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPUSH123200K400K600K800K1000KMin: 1244090.5 / Avg: 1247740 / Max: 1253938.62Min: 1230816.5 / Avg: 1242294.79 / Max: 1260263.62Min: 1245346 / Avg: 1249824 / Max: 1254882.751. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b0123246810SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.01, N = 38.288.308.33MIN: 8.18 / MAX: 8.87MIN: 8.14 / MAX: 8.79MIN: 8.16 / MAX: 9.171. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: efficientnet-b01233691215Min: 8.24 / Avg: 8.28 / Max: 8.3Min: 8.27 / Avg: 8.3 / Max: 8.33Min: 8.32 / Avg: 8.33 / Max: 8.341. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: Decompression123110220330440550SE +/- 3.00, N = 3SE +/- 0.67, N = 35035065051. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: Decompression12390180270360450Min: 497 / Avg: 503 / Max: 506Min: 504 / Avg: 505.33 / Max: 5061. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 101230.53441.06881.60322.13762.672SE +/- 0.011, N = 3SE +/- 0.026, N = 3SE +/- 0.013, N = 32.3752.3612.375
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 10123246810Min: 2.36 / Avg: 2.38 / Max: 2.39Min: 2.32 / Avg: 2.36 / Max: 2.41Min: 2.36 / Avg: 2.38 / Max: 2.4

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU1235001000150020002500SE +/- 0.62, N = 3SE +/- 1.77, N = 3SE +/- 9.79, N = 32208.252211.702221.13MIN: 2204.85MIN: 2206.98MIN: 2207.171. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPU123400800120016002000Min: 2207.04 / Avg: 2208.25 / Max: 2209.12Min: 2209.14 / Avg: 2211.7 / Max: 2215.1Min: 2209.7 / Avg: 2221.13 / Max: 2240.611. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.112370140210280350SE +/- 0.32, N = 3SE +/- 0.16, N = 3SE +/- 0.61, N = 3329.72328.63327.83MIN: 329.09 / MAX: 331.08MIN: 328.14 / MAX: 330.14MIN: 326.73 / MAX: 330.311. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.112360120180240300Min: 329.3 / Avg: 329.72 / Max: 330.36Min: 328.34 / Avg: 328.63 / Max: 328.87Min: 326.98 / Avg: 327.83 / Max: 329.011. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: LBC, LBRY Credits1235K10K15K20K25KSE +/- 113.58, N = 3SE +/- 86.86, N = 3SE +/- 96.84, N = 32333023407234631. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: LBC, LBRY Credits1234K8K12K16K20KMin: 23120 / Avg: 23330 / Max: 23510Min: 23310 / Avg: 23406.67 / Max: 23580Min: 23270 / Avg: 23463.33 / Max: 235701. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v2123246810SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 37.057.077.03MIN: 6.99 / MAX: 7.63MIN: 6.95 / MAX: 7.92MIN: 6.94 / MAX: 7.651. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: shufflenet-v21233691215Min: 7.03 / Avg: 7.05 / Max: 7.06Min: 7 / Avg: 7.07 / Max: 7.17Min: 6.98 / Avg: 7.03 / Max: 7.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU1235001000150020002500SE +/- 1.96, N = 3SE +/- 4.53, N = 3SE +/- 8.06, N = 32209.392214.442221.93MIN: 2203.59MIN: 2206.84MIN: 2205.081. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPU123400800120016002000Min: 2206.43 / Avg: 2209.39 / Max: 2213.09Min: 2209.5 / Avg: 2214.44 / Max: 2223.49Min: 2210.6 / Avg: 2221.93 / Max: 2237.521. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: x25x12350100150200250SE +/- 1.24, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3219.58220.73220.821. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: x25x1234080120160200Min: 217.1 / Avg: 219.58 / Max: 220.82Min: 220.66 / Avg: 220.73 / Max: 220.86Min: 220.68 / Avg: 220.82 / Max: 220.931. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU123246810SE +/- 0.02331, N = 3SE +/- 0.00084, N = 3SE +/- 0.03820, N = 37.416917.390277.43079MIN: 7.36MIN: 7.35MIN: 7.351. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: f32 - Engine: CPU1233691215Min: 7.39 / Avg: 7.42 / Max: 7.46Min: 7.39 / Avg: 7.39 / Max: 7.39Min: 7.39 / Avg: 7.43 / Max: 7.511. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Encryption12380160240320400SE +/- 1.43, N = 3SE +/- 0.34, N = 3SE +/- 0.07, N = 3344.4345.4346.2
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Encryption12360120180240300Min: 341.5 / Avg: 344.37 / Max: 345.9Min: 345 / Avg: 345.43 / Max: 346.1Min: 346.1 / Avg: 346.17 / Max: 346.3

OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-whirlpool123110K220K330K440K550KSE +/- 2447.10, N = 3SE +/- 541.00, N = 3SE +/- 720.67, N = 3530142532814532634
OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-whirlpool12390K180K270K360K450KMin: 525338 / Avg: 530141.67 / Max: 533355Min: 531732 / Avg: 532814 / Max: 533355Min: 531193 / Avg: 532634.33 / Max: 533355

TNN

TNN is an open-source deep learning reasoning framework developed by Tencent. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v212370140210280350SE +/- 1.18, N = 3SE +/- 0.24, N = 3SE +/- 0.90, N = 3339.21337.51338.74MIN: 336.01 / MAX: 348.17MIN: 336.44 / MAX: 346.96MIN: 336.55 / MAX: 341.551. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl
OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v212360120180240300Min: 336.86 / Avg: 339.21 / Max: 340.61Min: 337.07 / Avg: 337.51 / Max: 337.92Min: 337.38 / Avg: 338.74 / Max: 340.451. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: shufflenet-v2-10 - Device: OpenMP CPU1232K4K6K8K10KSE +/- 9.85, N = 3SE +/- 16.03, N = 3SE +/- 28.57, N = 39736977597291. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: shufflenet-v2-10 - Device: OpenMP CPU1232K4K6K8K10KMin: 9720 / Avg: 9736.17 / Max: 9754Min: 9747.5 / Avg: 9774.67 / Max: 9803Min: 9675.5 / Avg: 9728.5 / Max: 9773.51. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 61230.24080.48160.72240.96321.204SE +/- 0.008, N = 3SE +/- 0.003, N = 3SE +/- 0.002, N = 31.0651.0701.068
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 6123246810Min: 1.05 / Avg: 1.07 / Max: 1.08Min: 1.07 / Avg: 1.07 / Max: 1.08Min: 1.06 / Avg: 1.07 / Max: 1.07

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU1230.67591.35182.02772.70363.3795SE +/- 0.00849, N = 3SE +/- 0.01863, N = 3SE +/- 0.00263, N = 33.003892.990802.99407MIN: 2.94MIN: 2.83MIN: 2.941. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: u8s8f32 - Engine: CPU123246810Min: 2.99 / Avg: 3 / Max: 3.02Min: 2.95 / Avg: 2.99 / Max: 3.01Min: 2.99 / Avg: 2.99 / Max: 31. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Encryption123120240360480600SE +/- 2.47, N = 3SE +/- 1.50, N = 3SE +/- 0.07, N = 3549.0550.3551.4
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Encryption123100200300400500Min: 544.1 / Avg: 548.97 / Max: 552.1Min: 547.3 / Avg: 550.3 / Max: 552Min: 551.3 / Avg: 551.37 / Max: 551.5

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU1230.56021.12041.68062.24082.801SE +/- 0.00663, N = 3SE +/- 0.00118, N = 3SE +/- 0.00042, N = 32.489942.479302.48125MIN: 2.45MIN: 2.45MIN: 2.451. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPU123246810Min: 2.48 / Avg: 2.49 / Max: 2.5Min: 2.48 / Avg: 2.48 / Max: 2.48Min: 2.48 / Avg: 2.48 / Max: 2.481. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

WavPack Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPack12348121620SE +/- 0.09, N = 5SE +/- 0.02, N = 5SE +/- 0.03, N = 517.4017.3217.331. (CXX) g++ options: -rdynamic
OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPack12348121620Min: 17.29 / Avg: 17.39 / Max: 17.77Min: 17.29 / Avg: 17.32 / Max: 17.42Min: 17.29 / Avg: 17.33 / Max: 17.451. (CXX) g++ options: -rdynamic

Gcrypt Library

Libgcrypt is a general purpose cryptographic library developed as part of the GnuPG project. This is a benchmark of libgcrypt's integrated benchmark and is measuring the time to run the benchmark command with a cipher/mac/hash repetition count set for 50 times as simple, high level look at the overall crypto performance of the system under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterGcrypt Library 1.912360120180240300SE +/- 0.82, N = 3SE +/- 0.33, N = 3SE +/- 0.41, N = 3273.55272.41273.011. (CC) gcc options: -O2 -fvisibility=hidden
OpenBenchmarking.orgSeconds, Fewer Is BetterGcrypt Library 1.912350100150200250Min: 272.07 / Avg: 273.55 / Max: 274.91Min: 272.02 / Avg: 272.41 / Max: 273.07Min: 272.2 / Avg: 273.01 / Max: 273.551. (CC) gcc options: -O2 -fvisibility=hidden

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 0.8.1Video Input: Chimera 1080p 10-bit1231530456075SE +/- 0.18, N = 3SE +/- 0.17, N = 3SE +/- 0.08, N = 368.4468.6268.34MIN: 43.96 / MAX: 169.86MIN: 44.12 / MAX: 171.68MIN: 44.14 / MAX: 168.931. (CC) gcc options: -pthread
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Chimera 1080p 10-bit1231326395265Min: 68.13 / Avg: 68.44 / Max: 68.75Min: 68.39 / Avg: 68.62 / Max: 68.95Min: 68.19 / Avg: 68.34 / Max: 68.481. (CC) gcc options: -pthread

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC112350100150200250SE +/- 0.79, N = 3SE +/- 0.55, N = 3SE +/- 0.05, N = 3235.54235.97236.451. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC11234080120160200Min: 233.99 / Avg: 235.54 / Max: 236.56Min: 234.88 / Avg: 235.97 / Max: 236.52Min: 236.39 / Avg: 236.45 / Max: 236.551. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Decryption123120240360480600SE +/- 0.78, N = 3SE +/- 1.18, N = 3SE +/- 0.81, N = 3532.9531.5533.5
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Decryption12390180270360450Min: 531.5 / Avg: 532.93 / Max: 534.2Min: 529.9 / Avg: 531.5 / Max: 533.8Min: 531.9 / Avg: 533.47 / Max: 534.6

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC2123306090120150SE +/- 0.03, N = 3SE +/- 0.38, N = 3SE +/- 0.53, N = 3140.73140.32140.211. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC2123306090120150Min: 140.68 / Avg: 140.73 / Max: 140.76Min: 139.55 / Avg: 140.31 / Max: 140.74Min: 139.16 / Avg: 140.21 / Max: 140.751. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v31231.22182.44363.66544.88726.109SE +/- 0.05, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 35.415.415.43MIN: 5.29 / MAX: 5.77MIN: 5.29 / MAX: 5.54MIN: 5.28 / MAX: 6.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU-v3-v3 - Model: mobilenet-v3123246810Min: 5.34 / Avg: 5.41 / Max: 5.51Min: 5.37 / Avg: 5.41 / Max: 5.43Min: 5.37 / Avg: 5.43 / Max: 5.51. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg161231224364860SE +/- 0.13, N = 3SE +/- 0.08, N = 3SE +/- 0.06, N = 351.7851.9751.85MIN: 51.37 / MAX: 99.9MIN: 51.54 / MAX: 54.71MIN: 51.49 / MAX: 53.931. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: vgg161231020304050Min: 51.58 / Avg: 51.78 / Max: 52.03Min: 51.89 / Avg: 51.97 / Max: 52.12Min: 51.76 / Avg: 51.85 / Max: 51.971. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 11230.06230.12460.18690.24920.3115SE +/- 0.000, N = 3SE +/- 0.000, N = 3SE +/- 0.001, N = 30.2770.2770.276
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 112312345Min: 0.28 / Avg: 0.28 / Max: 0.28Min: 0.28 / Avg: 0.28 / Max: 0.28Min: 0.28 / Avg: 0.28 / Max: 0.28

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: DXT11232004006008001000SE +/- 2.71, N = 3SE +/- 0.37, N = 3SE +/- 2.11, N = 31083.861080.551084.331. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: DXT11232004006008001000Min: 1081.08 / Avg: 1083.86 / Max: 1089.28Min: 1079.82 / Avg: 1080.55 / Max: 1081.03Min: 1081.18 / Avg: 1084.33 / Max: 1088.341. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Decryption12330060090012001500SE +/- 1.87, N = 3SE +/- 2.11, N = 3SE +/- 2.16, N = 31389.61393.01394.3
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Decryption1232004006008001000Min: 1386 / Avg: 1389.6 / Max: 1392.3Min: 1389.9 / Avg: 1392.97 / Max: 1397Min: 1391.1 / Avg: 1394.27 / Max: 1398.4

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnet1233691215SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 312.0812.1212.11MIN: 11.98 / MAX: 12.56MIN: 12.02 / MAX: 12.63MIN: 12.03 / MAX: 12.391. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: alexnet12348121620Min: 12.05 / Avg: 12.08 / Max: 12.1Min: 12.08 / Avg: 12.12 / Max: 12.19Min: 12.1 / Avg: 12.11 / Max: 12.121. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - Gridding12330060090012001500SE +/- 0.36, N = 3SE +/- 0.31, N = 3SE +/- 0.57, N = 31285.021284.711280.901. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - Gridding1232004006008001000Min: 1284.4 / Avg: 1285.02 / Max: 1285.64Min: 1284.09 / Avg: 1284.71 / Max: 1285.02Min: 1279.77 / Avg: 1280.9 / Max: 1281.621. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Decryption12380160240320400SE +/- 0.25, N = 3SE +/- 0.20, N = 3SE +/- 1.02, N = 3346.4346.8345.7
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Decryption12360120180240300Min: 345.9 / Avg: 346.4 / Max: 346.7Min: 346.4 / Avg: 346.77 / Max: 347.1Min: 343.7 / Avg: 345.73 / Max: 346.9

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Decryption12380160240320400SE +/- 0.24, N = 3SE +/- 0.91, N = 3SE +/- 0.15, N = 3346.8345.8346.9
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Decryption12360120180240300Min: 346.3 / Avg: 346.77 / Max: 347.1Min: 344 / Avg: 345.77 / Max: 347Min: 346.7 / Avg: 346.93 / Max: 347.2

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPU12370140210280350SE +/- 1.01, N = 3SE +/- 0.60, N = 3SE +/- 0.44, N = 33283283271. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: yolov4 - Device: OpenMP CPU12360120180240300Min: 326 / Avg: 327.83 / Max: 329.5Min: 326.5 / Avg: 327.67 / Max: 328.5Min: 326.5 / Avg: 327.33 / Max: 3281. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: Decompression12330060090012001500SE +/- 2.08, N = 3SE +/- 3.06, N = 3SE +/- 2.73, N = 31390138913861. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: Decompression1232004006008001000Min: 1386 / Avg: 1390 / Max: 1393Min: 1383 / Avg: 1389 / Max: 1393Min: 1382 / Avg: 1385.67 / Max: 13911. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Encryption12330060090012001500SE +/- 1.19, N = 3SE +/- 3.43, N = 3SE +/- 0.72, N = 31400.21401.71404.2
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Encryption1232004006008001000Min: 1397.9 / Avg: 1400.2 / Max: 1401.9Min: 1395.4 / Avg: 1401.67 / Max: 1407.2Min: 1403 / Avg: 1404.23 / Max: 1405.5

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1239001800270036004500SE +/- 1.90, N = 3SE +/- 1.11, N = 3SE +/- 6.05, N = 34051.994040.594048.11MIN: 4039.27MIN: 4035.29MIN: 40331. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPU1237001400210028003500Min: 4049.57 / Avg: 4051.99 / Max: 4055.73Min: 4039.24 / Avg: 4040.59 / Max: 4042.79Min: 4036.86 / Avg: 4048.11 / Max: 4057.591. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: Compression12380160240320400SE +/- 0.33, N = 3SE +/- 0.33, N = 33563563571. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 0 - Process: Compression12360120180240300Min: 355 / Avg: 355.67 / Max: 356Min: 356 / Avg: 356.33 / Max: 3571. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU1230.7581.5162.2743.0323.79SE +/- 0.01155, N = 3SE +/- 0.00051, N = 3SE +/- 0.01022, N = 33.368503.359473.36883MIN: 3.3MIN: 3.3MIN: 3.271. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Matrix Multiply Batch Shapes Transformer - Data Type: f32 - Engine: CPU123246810Min: 3.35 / Avg: 3.37 / Max: 3.39Min: 3.36 / Avg: 3.36 / Max: 3.36Min: 3.36 / Avg: 3.37 / Max: 3.391. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU1230.63871.27741.91612.55483.1935SE +/- 0.00327, N = 3SE +/- 0.00201, N = 3SE +/- 0.00573, N = 32.830742.838612.83373MIN: 2.8MIN: 2.8MIN: 2.791. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPU123246810Min: 2.82 / Avg: 2.83 / Max: 2.84Min: 2.83 / Avg: 2.84 / Max: 2.84Min: 2.82 / Avg: 2.83 / Max: 2.841. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -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 0.8.1Video Input: Chimera 1080p123100200300400500SE +/- 0.72, N = 3SE +/- 0.88, N = 3SE +/- 2.60, N = 3459.11459.32458.06MIN: 341.89 / MAX: 588.95MIN: 342.72 / MAX: 583.61MIN: 341.78 / MAX: 587.391. (CC) gcc options: -pthread
OpenBenchmarking.orgFPS, More Is Betterdav1d 0.8.1Video Input: Chimera 1080p12380160240320400Min: 457.68 / Avg: 459.11 / Max: 460.01Min: 457.57 / Avg: 459.32 / Max: 460.27Min: 452.87 / Avg: 458.06 / Max: 460.841. (CC) gcc options: -pthread

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: SqueezeNetV1.0123246810SE +/- 0.011, N = 3SE +/- 0.029, N = 3SE +/- 0.021, N = 38.4478.4398.424MIN: 8.35 / MAX: 9.56MIN: 8.3 / MAX: 52.09MIN: 8.33 / MAX: 9.341. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: SqueezeNetV1.01233691215Min: 8.43 / Avg: 8.45 / Max: 8.47Min: 8.4 / Avg: 8.44 / Max: 8.5Min: 8.4 / Avg: 8.42 / Max: 8.471. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: Compression12390180270360450SE +/- 0.58, N = 3SE +/- 1.00, N = 34014014001. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 1 - Process: Compression12370140210280350Min: 400 / Avg: 401 / Max: 402Min: 398 / Avg: 400 / Max: 4011. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

rav1e

Xiph rav1e is a Rust-written AV1 video encoder. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 51230.18230.36460.54690.72920.9115SE +/- 0.001, N = 3SE +/- 0.005, N = 3SE +/- 0.003, N = 30.8100.8100.808
OpenBenchmarking.orgFrames Per Second, More Is Betterrav1e 0.4Speed: 5123246810Min: 0.81 / Avg: 0.81 / Max: 0.81Min: 0.8 / Avg: 0.81 / Max: 0.82Min: 0.8 / Avg: 0.81 / Max: 0.81

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 95, Compression Effort 7123120240360480600SE +/- 1.36, N = 3SE +/- 0.47, N = 3SE +/- 0.96, N = 3545.96545.66544.661. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg
OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 95, Compression Effort 7123100200300400500Min: 543.3 / Avg: 545.96 / Max: 547.79Min: 544.84 / Avg: 545.66 / Max: 546.45Min: 542.76 / Avg: 544.66 / Max: 545.821. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg

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 30M12350100150200250SE +/- 0.54, N = 3SE +/- 0.61, N = 3SE +/- 1.02, N = 3219.05219.10219.571. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -lgenericPatchFields -lOpenFOAM -ldl -lm
OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30M1234080120160200Min: 218.14 / Avg: 219.05 / Max: 220.02Min: 218.32 / Avg: 219.1 / Max: 220.3Min: 217.59 / Avg: 219.57 / Max: 2211. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -ldynamicMesh -lgenericPatchFields -lOpenFOAM -ldl -lm

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: Decompression123901802703604504314324311. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Google SynthMark

SynthMark is a cross platform tool for benchmarking CPU performance under a variety of real-time audio workloads. It uses a polyphonic synthesizer model to provide standardized tests for latency, jitter and computational throughput. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100123120240360480600SE +/- 0.84, N = 3SE +/- 0.57, N = 3SE +/- 0.80, N = 3551.18550.14549.941. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast
OpenBenchmarking.orgVoices, More Is BetterGoogle SynthMark 20201109Test: VoiceMark_100123100200300400500Min: 549.57 / Avg: 551.18 / Max: 552.42Min: 549.49 / Avg: 550.14 / Max: 551.27Min: 548.47 / Avg: 549.93 / Max: 551.231. (CXX) g++ options: -lm -lpthread -std=c++11 -Ofast

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU123246810SE +/- 0.00368, N = 3SE +/- 0.00711, N = 3SE +/- 0.00234, N = 38.393258.401248.38226MIN: 8.35MIN: 8.35MIN: 8.311. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPU1233691215Min: 8.39 / Avg: 8.39 / Max: 8.4Min: 8.39 / Avg: 8.4 / Max: 8.41Min: 8.38 / Avg: 8.38 / Max: 8.381. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU1231.30012.60023.90035.20046.5005SE +/- 0.01646, N = 3SE +/- 0.00194, N = 3SE +/- 0.00979, N = 35.765365.778235.76890MIN: 5.7MIN: 5.74MIN: 5.711. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPU123246810Min: 5.75 / Avg: 5.77 / Max: 5.8Min: 5.77 / Avg: 5.78 / Max: 5.78Min: 5.75 / Avg: 5.77 / Max: 5.781. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 75, Compression Effort 712370140210280350SE +/- 0.36, N = 3SE +/- 0.19, N = 3SE +/- 0.52, N = 3298.92298.25298.541. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg
OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 75, Compression Effort 712350100150200250Min: 298.21 / Avg: 298.92 / Max: 299.39Min: 297.92 / Avg: 298.25 / Max: 298.57Min: 297.91 / Avg: 298.54 / Max: 299.581. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Decryption123120240360480600SE +/- 0.52, N = 3SE +/- 0.39, N = 3533.7533.3532.6
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Decryption12390180270360450Min: 532.3 / Avg: 533.27 / Max: 534.1Min: 531.8 / Avg: 532.57 / Max: 533.1

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU1233691215SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 312.7112.6912.72MIN: 12.64MIN: 12.63MIN: 12.641. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPU12348121620Min: 12.71 / Avg: 12.71 / Max: 12.71Min: 12.69 / Avg: 12.69 / Max: 12.7Min: 12.71 / Avg: 12.72 / Max: 12.751. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Encryption123120240360480600SE +/- 0.60, N = 2SE +/- 0.57, N = 3SE +/- 0.26, N = 3550.7551.2551.7
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Encryption123100200300400500Min: 550.1 / Avg: 550.7 / Max: 551.3Min: 550.1 / Avg: 551.23 / Max: 551.8Min: 551.2 / Avg: 551.67 / Max: 552.1

Mobile Neural Network

MNN is the Mobile Neural Network as a highly efficient, lightweight deep learning framework developed by Alibaba. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: inception-v31231326395265SE +/- 0.19, N = 3SE +/- 0.21, N = 3SE +/- 0.26, N = 355.8755.9255.96MIN: 55.38 / MAX: 87.21MIN: 55.44 / MAX: 148.38MIN: 55.41 / MAX: 122.631. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl
OpenBenchmarking.orgms, Fewer Is BetterMobile Neural Network 1.1.1Model: inception-v31231122334455Min: 55.49 / Avg: 55.87 / Max: 56.11Min: 55.6 / Avg: 55.92 / Max: 56.32Min: 55.56 / Avg: 55.96 / Max: 56.441. (CXX) g++ options: -std=c++11 -O3 -fvisibility=hidden -fomit-frame-pointer -fstrict-aliasing -ffunction-sections -fdata-sections -ffast-math -fno-rtti -fno-exceptions -rdynamic -pthread -ldl

Etcpak

Etcpack is the self-proclaimed "fastest ETC compressor on the planet" with focused on providing open-source, very fast ETC and S3 texture compression support. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1 + Dithering12350100150200250SE +/- 0.40, N = 3SE +/- 0.03, N = 3SE +/- 0.05, N = 3226.75227.14227.061. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread
OpenBenchmarking.orgMpx/s, More Is BetterEtcpak 0.7Configuration: ETC1 + Dithering1234080120160200Min: 225.94 / Avg: 226.75 / Max: 227.18Min: 227.11 / Avg: 227.14 / Max: 227.19Min: 226.95 / Avg: 227.06 / Max: 227.121. (CXX) g++ options: -O3 -march=native -std=c++11 -lpthread

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMP12350100150200250SE +/- 0.33, N = 3SE +/- 0.19, N = 3SE +/- 0.19, N = 3239.24239.43239.621. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgIterations Per Second, More Is BetterASKAP 1.0Test: Hogbom Clean OpenMP1234080120160200Min: 238.66 / Avg: 239.24 / Max: 239.81Min: 239.23 / Avg: 239.43 / Max: 239.81Min: 239.23 / Avg: 239.62 / Max: 239.811. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Timed Godot Game Engine Compilation

This test times how long it takes to compile the Godot Game Engine. Godot is a popular, open-source, cross-platform 2D/3D game engine and is built using the SCons build system and targeting the X11 platform. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To Compile1234080120160200SE +/- 0.10, N = 3SE +/- 0.26, N = 3SE +/- 0.38, N = 3186.70186.84186.56
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To Compile123306090120150Min: 186.53 / Avg: 186.7 / Max: 186.89Min: 186.38 / Avg: 186.84 / Max: 187.26Min: 185.89 / Avg: 186.56 / Max: 187.22

NCNN

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

OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssd123612182430SE +/- 0.04, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 326.5526.5726.59MIN: 26.01 / MAX: 27.38MIN: 26.04 / MAX: 27.24MIN: 26.05 / MAX: 27.211. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread
OpenBenchmarking.orgms, Fewer Is BetterNCNN 20201218Target: CPU - Model: squeezenet_ssd123612182430Min: 26.5 / Avg: 26.55 / Max: 26.63Min: 26.52 / Avg: 26.57 / Max: 26.59Min: 26.54 / Avg: 26.59 / Max: 26.661. (CXX) g++ options: -O3 -rdynamic -lgomp -lpthread

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Lossless Compression1232004006008001000SE +/- 0.20, N = 3SE +/- 0.53, N = 3SE +/- 0.05, N = 3939.76938.92940.231. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg
OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Lossless Compression123160320480640800Min: 939.45 / Avg: 939.76 / Max: 940.14Min: 938.09 / Avg: 938.92 / Max: 939.9Min: 940.16 / Avg: 940.23 / Max: 940.341. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU1231.14242.28483.42724.56965.712SE +/- 0.00123, N = 3SE +/- 0.00166, N = 3SE +/- 0.00135, N = 35.075645.077145.07005MIN: 5.05MIN: 5.06MIN: 5.041. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPU123246810Min: 5.07 / Avg: 5.08 / Max: 5.08Min: 5.08 / Avg: 5.08 / Max: 5.08Min: 5.07 / Avg: 5.07 / Max: 5.071. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU12348121620SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 313.7913.8013.78MIN: 13.69MIN: 13.72MIN: 13.711. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPU12348121620Min: 13.78 / Avg: 13.79 / Max: 13.8Min: 13.8 / Avg: 13.8 / Max: 13.8Min: 13.77 / Avg: 13.78 / Max: 13.791. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU1230.9181.8362.7543.6724.59SE +/- 0.00717, N = 3SE +/- 0.01870, N = 3SE +/- 0.02757, N = 34.074764.074514.07980MIN: 4MIN: 3.98MIN: 3.991. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: IP Shapes 1D - Data Type: f32 - Engine: CPU123246810Min: 4.06 / Avg: 4.07 / Max: 4.09Min: 4.05 / Avg: 4.07 / Max: 4.11Min: 4.04 / Avg: 4.08 / Max: 4.131. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU1239001800270036004500SE +/- 2.55, N = 3SE +/- 1.68, N = 3SE +/- 3.06, N = 34045.794044.154040.61MIN: 4037.43MIN: 4035.82MIN: 4031.11. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPU1237001400210028003500Min: 4040.74 / Avg: 4045.79 / Max: 4048.94Min: 4040.84 / Avg: 4044.15 / Max: 4046.28Min: 4034.69 / Avg: 4040.61 / Max: 4044.91. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

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 1.21123400800120016002000SE +/- 3.77, N = 3SE +/- 2.43, N = 3SE +/- 4.94, N = 31689.61691.51691.61. (CXX) g++ options: -O3 -march=native -rdynamic
OpenBenchmarking.orgMFLOPS, More Is BetterQuantLib 1.2112330060090012001500Min: 1682.3 / Avg: 1689.6 / Max: 1694.9Min: 1686.7 / Avg: 1691.53 / Max: 1694.3Min: 1683 / Avg: 1691.57 / Max: 1700.11. (CXX) g++ options: -O3 -march=native -rdynamic

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Decompression1232004006008001000SE +/- 0.33, N = 39959949951. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Decompression1232004006008001000Min: 994 / Avg: 994.33 / Max: 9951. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-sha512123300K600K900K1200K1500KSE +/- 562.33, N = 3132899313301181328993
OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-sha512123200K400K600K800K1000KMin: 1328993 / Avg: 1330117.67 / Max: 1330680

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPU1239001800270036004500SE +/- 9.80, N = 3SE +/- 5.93, N = 3SE +/- 8.12, N = 34007400840051. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: super-resolution-10 - Device: OpenMP CPU1237001400210028003500Min: 3992 / Avg: 4007.17 / Max: 4025.5Min: 3997 / Avg: 4008.33 / Max: 4017Min: 3995.5 / Avg: 4004.83 / Max: 40211. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1239001800270036004500SE +/- 5.71, N = 3SE +/- 6.88, N = 3SE +/- 5.47, N = 34048.954050.544051.42MIN: 4036MIN: 4036.39MIN: 4040.911. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPU1237001400210028003500Min: 4040.26 / Avg: 4048.95 / Max: 4059.72Min: 4042.55 / Avg: 4050.54 / Max: 4064.23Min: 4045.62 / Avg: 4051.42 / Max: 4062.341. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Compression Effort 512348121620SE +/- 0.02, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 314.4814.4814.481. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg
OpenBenchmarking.orgSeconds, Fewer Is BetterWebP2 Image Encode 20210126Encode Settings: Quality 100, Compression Effort 512348121620Min: 14.45 / Avg: 14.48 / Max: 14.52Min: 14.45 / Avg: 14.48 / Max: 14.52Min: 14.44 / Avg: 14.47 / Max: 14.531. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -rdynamic -lpthread -ljpeg

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU1235001000150020002500SE +/- 5.90, N = 3SE +/- 0.63, N = 3SE +/- 2.02, N = 32211.562211.322210.61MIN: 2201.95MIN: 2206.12MIN: 2204.441. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread
OpenBenchmarking.orgms, Fewer Is BetteroneDNN 2.0Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPU123400800120016002000Min: 2204.26 / Avg: 2211.56 / Max: 2223.25Min: 2210.48 / Avg: 2211.32 / Max: 2212.55Min: 2206.81 / Avg: 2210.61 / Max: 2213.721. (CXX) g++ options: -O3 -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Encryption12380160240320400SE +/- 0.17, N = 3SE +/- 0.10, N = 3345.6345.7345.6
OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Encryption12360120180240300Min: 345.4 / Avg: 345.73 / Max: 345.9Min: 345.4 / Avg: 345.6 / Max: 345.7

ONNX Runtime

ONNX Runtime is developed by Microsoft and partners as a open-source, cross-platform, high performance machine learning inferencing and training accelerator. This test profile runs the ONNX Runtime with various models available from the ONNX Zoo. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: fcn-resnet101-11 - Device: OpenMP CPU1231326395265SE +/- 0.00, N = 3SE +/- 0.17, N = 35959591. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt
OpenBenchmarking.orgInferences Per Minute, More Is BetterONNX Runtime 1.6Model: fcn-resnet101-11 - Device: OpenMP CPU1231224364860Min: 58.5 / Avg: 58.5 / Max: 58.5Min: 58.5 / Avg: 58.83 / Max: 591. (CXX) g++ options: -fopenmp -ffunction-sections -fdata-sections -O3 -ldl -lrt

lzbench

lzbench is an in-memory benchmark of various compressors. The file used for compression is a Linux kernel source tree tarball. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Libdeflate 1 - Process: Compression12340801201602001821821821. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Brotli 2 - Process: Compression1233060901201501481481481. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Crush 0 - Process: Compression123204060801007676761. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: Compression1231530456075SE +/- 0.58, N = 36565651. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: Zstd 8 - Process: Compression1231326395265Min: 64 / Avg: 65 / Max: 661. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: XZ 0 - Process: Decompression123204060801009696961. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: XZ 0 - Process: Compression123816243240SE +/- 0.33, N = 33434341. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 1.8Test: XZ 0 - Process: Compression123714212835Min: 33 / Avg: 33.67 / Max: 341. (CXX) g++ options: -pthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

Redis

Redis is an open-source in-memory data structure store, used as a database, cache, and message broker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPOP123400K800K1200K1600K2000KSE +/- 6269.54, N = 3SE +/- 4394.55, N = 3SE +/- 91181.44, N = 122010121.551271416.381758555.231. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3
OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 6.0.9Test: LPOP123300K600K900K1200K1500KMin: 1997602.88 / Avg: 2010121.55 / Max: 2017000.38Min: 1262801.88 / Avg: 1271416.38 / Max: 1277233.25Min: 1199048.38 / Avg: 1758555.23 / Max: 1978630.881. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3

ASKAP

ASKAP is a set of benchmarks from the Australian SKA Pathfinder. The principal ASKAP benchmarks are the Hogbom Clean Benchmark (tHogbomClean) and Convolutional Resamping Benchmark (tConvolve) as well as some previous ASKAP benchmarks being included as well for OpenCL and CUDA execution of tConvolve. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - Gridding123400800120016002000SE +/- 39.18, N = 15SE +/- 31.45, N = 15SE +/- 12.81, N = 31778.101881.941962.741. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - Gridding12330060090012001500Min: 1633.47 / Avg: 1778.1 / Max: 1972.27Min: 1664.1 / Avg: 1881.94 / Max: 1986.99Min: 1943.47 / Avg: 1962.74 / Max: 1986.991. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - Gridding123400800120016002000SE +/- 65.46, N = 3SE +/- 45.52, N = 15SE +/- 47.93, N = 122071.461984.731979.001. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - Gridding123400800120016002000Min: 1940.8 / Avg: 2071.46 / Max: 2143.76Min: 1631.81 / Avg: 1984.73 / Max: 2246.54Min: 1699.45 / Avg: 1979 / Max: 2201.311. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - Degridding123400800120016002000SE +/- 18.20, N = 3SE +/- 28.57, N = 15SE +/- 29.00, N = 121699.851589.751597.841. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp
OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - Degridding12330060090012001500Min: 1664.95 / Avg: 1699.85 / Max: 1726.29Min: 1395.72 / Avg: 1589.75 / Max: 1782.58Min: 1457.75 / Avg: 1597.84 / Max: 1812.121. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

Cpuminer-Opt

Cpuminer-Opt is a fork of cpuminer-multi that carries a wide range of CPU performance optimizations for measuring the potential cryptocurrency mining performance of the CPU/processor with a wide variety of cryptocurrencies. The benchmark reports the hash speed for the CPU mining performance for the selected cryptocurrency. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Triple SHA-256, Onecoin12313K26K39K52K65KSE +/- 902.62, N = 15SE +/- 1076.82, N = 15SE +/- 80.07, N = 36124760350629231. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Triple SHA-256, Onecoin12311K22K33K44K55KMin: 53730 / Avg: 61246.67 / Max: 66670Min: 53020 / Avg: 60350 / Max: 66780Min: 62770 / Avg: 62923.33 / Max: 630401. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Quad SHA-256, Pyrite12310K20K30K40K50KSE +/- 749.50, N = 3SE +/- 748.75, N = 15SE +/- 1271.64, N = 154765348622483411. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Quad SHA-256, Pyrite1238K16K24K32K40KMin: 46220 / Avg: 47653.33 / Max: 48750Min: 45320 / Avg: 48622 / Max: 55630Min: 33410 / Avg: 48340.67 / Max: 537101. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Deepcoin12316003200480064008000SE +/- 132.89, N = 15SE +/- 164.38, N = 15SE +/- 33.20, N = 37487.987013.877312.241. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp
OpenBenchmarking.orgkH/s, More Is BetterCpuminer-Opt 3.15.5Algorithm: Deepcoin12313002600390052006500Min: 7198.21 / Avg: 7487.98 / Max: 9313.84Min: 5736.81 / Avg: 7013.87 / Max: 7574.65Min: 7247.16 / Avg: 7312.24 / Max: 7356.161. (CXX) g++ options: -O2 -lcurl -lz -lpthread -lssl -lcrypto -lgmp

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin Protein1231.00422.00843.01264.01685.021SE +/- 0.131, N = 15SE +/- 0.017, N = 3SE +/- 0.121, N = 154.4154.2614.4631. (CXX) g++ options: -O3 -pthread -lm
OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin Protein123246810Min: 3.75 / Avg: 4.41 / Max: 5.35Min: 4.23 / Avg: 4.26 / Max: 4.29Min: 3.55 / Avg: 4.46 / Max: 5.11. (CXX) g++ options: -O3 -pthread -lm

CLOMP

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

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP Speedup1233691215SE +/- 0.13, N = 3SE +/- 1.35, N = 12SE +/- 0.09, N = 313.211.013.01. (CC) gcc options: -fopenmp -O3 -lm
OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 1.2Static OMP Speedup12348121620Min: 12.9 / Avg: 13.17 / Max: 13.3Min: 1 / Avg: 11.04 / Max: 13.1Min: 12.9 / Avg: 13.03 / Max: 13.21. (CC) gcc options: -fopenmp -O3 -lm

137 Results Shown

Redis
NAS Parallel Benchmarks
NCNN
Cpuminer-Opt
CloverLeaf
FinanceBench
oneDNN
Cpuminer-Opt
NCNN
FinanceBench
CP2K Molecular Dynamics
NCNN
Redis
ASKAP
Cpuminer-Opt
lzbench
NAS Parallel Benchmarks
Opus Codec Encoding
Mobile Neural Network
WebP2 Image Encode
Kripke
QMCPACK
NCNN:
  CPU - blazeface
  CPU-v2-v2 - mobilenet-v2
  CPU - regnety_400m
Mobile Neural Network
dav1d
GnuPG
Timed Eigen Compilation
dav1d
Cpuminer-Opt
Redis
Mobile Neural Network
perf-bench
Unpacking Firefox
NCNN:
  CPU - mobilenet
  CPU - resnet50
Cpuminer-Opt
Cryptsetup
Algebraic Multi-Grid Benchmark
ONNX Runtime
NCNN
Cryptsetup
LULESH
Cpuminer-Opt
lzbench
Build2
ASKAP
Redis
NCNN
lzbench
rav1e
oneDNN
TNN
Cpuminer-Opt
NCNN
oneDNN
Cpuminer-Opt
oneDNN
Cryptsetup:
  Twofish-XTS 256b Encryption
  PBKDF2-whirlpool
TNN
ONNX Runtime
rav1e
oneDNN
Cryptsetup
oneDNN
WavPack Audio Encoding
Gcrypt Library
dav1d
Etcpak
Cryptsetup
Etcpak
NCNN:
  CPU-v3-v3 - mobilenet-v3
  CPU - vgg16
rav1e
Etcpak
Cryptsetup
NCNN
ASKAP
Cryptsetup:
  Twofish-XTS 512b Decryption
  Twofish-XTS 256b Decryption
ONNX Runtime
lzbench
Cryptsetup
oneDNN
lzbench
oneDNN:
  Matrix Multiply Batch Shapes Transformer - f32 - CPU
  IP Shapes 1D - u8s8f32 - CPU
dav1d
Mobile Neural Network
lzbench
rav1e
WebP2 Image Encode
OpenFOAM
lzbench
Google SynthMark
oneDNN:
  Deconvolution Batch shapes_3d - f32 - CPU
  Deconvolution Batch shapes_1d - f32 - CPU
WebP2 Image Encode
Cryptsetup
oneDNN
Cryptsetup
Mobile Neural Network
Etcpak
ASKAP
Timed Godot Game Engine Compilation
NCNN
WebP2 Image Encode
oneDNN:
  Deconvolution Batch shapes_3d - u8s8f32 - CPU
  Convolution Batch Shapes Auto - f32 - CPU
  IP Shapes 1D - f32 - CPU
  Recurrent Neural Network Training - u8s8f32 - CPU
QuantLib
lzbench
Cryptsetup
ONNX Runtime
oneDNN
WebP2 Image Encode
oneDNN
Cryptsetup
ONNX Runtime
lzbench:
  Libdeflate 1 - Compression
  Brotli 2 - Compression
  Crush 0 - Compression
  Zstd 8 - Compression
  XZ 0 - Decompression
  XZ 0 - Compression
Redis
ASKAP:
  tConvolve OpenMP - Gridding
  tConvolve MPI - Gridding
  tConvolve MPI - Degridding
Cpuminer-Opt:
  Triple SHA-256, Onecoin
  Quad SHA-256, Pyrite
  Deepcoin
LAMMPS Molecular Dynamics Simulator
CLOMP

1

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
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x44
Python Notes: Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 26 January 2021 16:22 by user phoronix.

2

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
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x44
Python Notes: Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 27 January 2021 05:17 by user phoronix.

3

Processor: Intel Xeon E5-2687W v3 @ 3.50GHz (10 Cores / 20 Threads), Motherboard: MSI X99S SLI PLUS (MS-7885) v1.0 (1.E0 BIOS), Chipset: Intel Xeon E7 v3/Xeon, Memory: 32GB, Disk: 80GB INTEL SSDSCKGW08, Graphics: NVIDIA GeForce GTX 770, Audio: Realtek ALC892, Monitor: LG Ultra HD, Network: Intel I218-V

OS: Ubuntu 20.04, Kernel: 5.9.0-050900rc7daily20200928-generic (x86_64) 20200927, Desktop: GNOME Shell 3.36.4, Display Server: X Server 1.20.8, Display Driver: modesetting 1.20.8, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 3840x2160

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
Processor Notes: Scaling Governor: intel_cpufreq ondemand - CPU Microcode: 0x44
Python Notes: Python 3.8.5
Security Notes: itlb_multihit: KVM: Mitigation of VMX disabled + l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 27 January 2021 15:34 by user phoronix.