new tests extra cpus

Tests for a future article. Intel Xeon Gold 6346 testing with a Supermicro X12SPO-NTF v2.00 (1.2 BIOS) and astdrmfb on AlmaLinux 9.1 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 2304246-NE-NEWTESTSE15
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
Show Result Confidence Charts

Limit displaying results to tests within:

Timed Code Compilation 6 Tests
C/C++ Compiler Tests 8 Tests
CPU Massive 13 Tests
Creator Workloads 9 Tests
Cryptography 2 Tests
Database Test Suite 3 Tests
Encoding 4 Tests
Fortran Tests 3 Tests
Game Development 3 Tests
HPC - High Performance Computing 5 Tests
Common Kernel Benchmarks 4 Tests
Machine Learning 3 Tests
Multi-Core 15 Tests
Intel oneAPI 2 Tests
OpenMPI Tests 2 Tests
Programmer / Developer System Benchmarks 6 Tests
Python Tests 6 Tests
Server 6 Tests
Server CPU Tests 10 Tests
Video Encoding 3 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

Additional Graphs

Show Perf Per Core/Thread Calculation Graphs Where Applicable
Show Perf Per Clock Calculation Graphs Where Applicable

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results 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
e
April 04 2023
  5 Hours, 33 Minutes
f
April 04 2023
  5 Hours, 32 Minutes
g
April 04 2023
  2 Hours, 7 Minutes
h
April 22 2023
  1 Day, 8 Hours, 40 Minutes
i
April 23 2023
  11 Hours, 28 Minutes
j
April 23 2023
  11 Hours, 30 Minutes
Invert Hiding All Results Option
  11 Hours, 28 Minutes

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):


new tests extra cpusOpenBenchmarking.orgPhoronix Test SuiteIntel Xeon Gold 6338 @ 3.20GHz (32 Cores / 64 Threads)Intel Xeon Gold 6346 @ 3.60GHz (16 Cores / 32 Threads)Supermicro X12SPO-NTF v2.00 (1.2 BIOS)8 x 64 GB DDR4-3200MT/s Samsung M393A8G40AB2-CWE2 x 1920GB SAMSUNG MZQL21T9HCJR-00A07astdrmfbDELL E207WFPAlmaLinux 9.15.14.0-162.12.1.el9_1.x86_64 (x86_64)GCC 11.3.1 20220421ext41280x1024ProcessorsMotherboardMemoryDiskGraphicsMonitorOSKernelCompilerFile-SystemScreen Resolution New Tests Extra Cpus PerformanceSystem Logs- Transparent Huge Pages: always- --build=x86_64-redhat-linux --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-cet --enable-checking=release --enable-gnu-indirect-function --enable-gnu-unique-object --enable-host-bind-now --enable-host-pie --enable-initfini-array --enable-languages=c,c++,fortran,lto --enable-link-serialization=1 --enable-multilib --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=x86-64 --with-arch_64=x86-64-v2 --with-build-config=bootstrap-lto --with-gcc-major-version-only --with-linker-hash-style=gnu --with-tune=generic --without-cuda-driver --without-isl - Scaling Governor: intel_cpufreq performance - CPU Microcode: 0xd000375- Python 3.9.14- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Vulnerable: Clear buffers attempted no microcode; SMT vulnerable + retbleed: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

efghijResult OverviewPhoronix Test Suite100%114%127%141%Embree7-Zip CompressionOpenSSLJohn The RipperDarmstadt Automotive Parallel Heterogeneous SuiteTimed Linux Kernel CompilationStress-NGTimed FFmpeg CompilationRocksDBFFmpegSVT-AV1MariaDBGoogle Draco

new tests extra cpusnwchem: C240 Buckyballdeeprec: PLE - BF16deeprec: PLE - BF16deeprec: MMOE - BF16deeprec: MMOE - BF16deeprec: PLE - FP32deeprec: PLE - FP32tensorflow: CPU - 512 - ResNet-50deeprec: BST - BF16deeprec: BST - BF16deeprec: DCNv2 - BF16deeprec: DCNv2 - BF16blender: Barbershop - CPU-Onlydeeprec: MMOE - FP32deeprec: MMOE - FP32deeprec: DCNv2 - FP32deeprec: DCNv2 - FP32deeprec: BST - FP32deeprec: BST - FP32tensorflow: CPU - 256 - ResNet-50build-llvm: Unix Makefilesbuild-llvm: Ninjabuild-nodejs: Time To Compilepetsc: Streamstensorflow: CPU - 512 - GoogLeNetffmpeg: libx264 - Uploadffmpeg: libx264 - Uploadblender: Pabellon Barcelona - CPU-Onlymysqlslap: 8192build-godot: Time To Compiledeeprec: DLRM - BF16deeprec: DLRM - BF16opencv: Stitchingblender: Classroom - CPU-Onlyffmpeg: libx264 - Video On Demandffmpeg: libx264 - Video On Demandffmpeg: libx264 - Platformffmpeg: libx264 - Platformffmpeg: libx265 - Uploadffmpeg: libx265 - Uploadffmpeg: libx265 - Video On Demandffmpeg: libx265 - Video On Demandffmpeg: libx265 - Platformffmpeg: libx265 - Platformmemcached: 1:100openssl: SHA256openssl: ChaCha20-Poly1305openssl: AES-256-GCMopenssl: AES-128-GCMopenssl: ChaCha20openssl: SHA512opencv: Graph APItensorflow: CPU - 256 - GoogLeNetmysqlslap: 512build2: Time To Compiletensorflow: CPU - 512 - AlexNetvvenc: Bosphorus 4K - Fasttensorflow: CPU - 64 - ResNet-50mysqlslap: 4096mysqlslap: 2048mysqlslap: 1024opencv: Image Processingmysqlslap: 256blender: Fishy Cat - CPU-Onlydeeprec: DLRM - FP32deeprec: DLRM - FP32nginx: 1000nginx: 500nginx: 200apache: 1000apache: 500nginx: 100apache: 200apache: 100stress-ng: Atomicstress-ng: NUMAstress-ng: MEMFDonednn: Recurrent Neural Network Training - u8s8f32 - CPUonednn: Recurrent Neural Network Training - f32 - CPUonednn: Recurrent Neural Network Training - bf16bf16bf16 - CPUblender: BMW27 - CPU-Onlyonednn: Recurrent Neural Network Inference - bf16bf16bf16 - CPUonednn: Recurrent Neural Network Inference - f32 - CPUonednn: Recurrent Neural Network Inference - u8s8f32 - CPUstress-ng: Futexstress-ng: Socket Activitystress-ng: MEMFDopencv: Corestress-ng: Futexopencv: Object Detectiontensorflow: CPU - 256 - AlexNettensorflow: CPU - 32 - ResNet-50memcached: 1:10daphne: OpenMP - Points2Imagememcached: 1:5vvenc: Bosphorus 4K - Fasterstress-ng: CPU Cacheffmpeg: libx265 - Liveffmpeg: libx265 - Livebuild-linux-kernel: defconfigjohn-the-ripper: MD5rocksdb: Rand Fillrocksdb: Update Randrocksdb: Rand Fill Syncrocksdb: Read Rand Write Randrocksdb: Read While Writingrocksdb: Rand Readjohn-the-ripper: HMAC-SHA512openssl: RSA4096openssl: RSA4096encode-opus: WAV To Opus Encodeopencv: Features 2Dsvt-av1: Preset 4 - Bosphorus 4Kdeeprec: DIN - BF16deeprec: DIN - BF16tensorflow: CPU - 64 - GoogLeNetvvenc: Bosphorus 1080p - Fastffmpeg: libx264 - Liveffmpeg: libx264 - Livedeeprec: DIN - FP32deeprec: DIN - FP32tensorflow: CPU - 16 - ResNet-50compress-7zip: Decompression Ratingcompress-7zip: Compression Ratingembree: Pathtracer - Asian Dragon Objopencv: DNN - Deep Neural Networkstress-ng: Context Switchingembree: Pathtracer ISPC - Asian Dragon Objbuild-ffmpeg: Time To Compilerocksdb: Seq Fillstress-ng: Zlibstress-ng: Forkingstress-ng: Pthreadstress-ng: Hashstress-ng: Semaphoresstress-ng: CPU Stressstress-ng: MMAPstress-ng: Vector Mathstress-ng: Matrix Mathstress-ng: Mallocstress-ng: CPU Cachestress-ng: Pollstress-ng: Cryptostress-ng: Mutexstress-ng: SENDFILEstress-ng: Zlibdaphne: OpenMP - Euclidean Clusterstress-ng: Atomicjohn-the-ripper: WPA PSKstress-ng: Pthreadstress-ng: NUMAstress-ng: Forkingstress-ng: System V Message Passingstress-ng: Semaphoresjohn-the-ripper: bcryptjohn-the-ripper: Blowfishstress-ng: Vector Mathstress-ng: Hashstress-ng: SENDFILEstress-ng: CPU Stressstress-ng: Mallocstress-ng: Memory Copyingstress-ng: Glibc Qsort Data Sortingstress-ng: Glibc C String Functionsstress-ng: Matrix Mathstress-ng: MMAPstress-ng: Function Callstress-ng: Mutexstress-ng: Pollstress-ng: Cryptostress-ng: x86_64 RdRandembree: Pathtracer - Crowntensorflow: CPU - 32 - GoogLeNetembree: Pathtracer ISPC - Crownembree: Pathtracer - Asian Dragonvvenc: Bosphorus 1080p - Fasteronednn: Deconvolution Batch shapes_1d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_1d - f32 - CPUonednn: Deconvolution Batch shapes_1d - u8s8f32 - CPUtensorflow: CPU - 64 - AlexNetsvt-av1: Preset 4 - Bosphorus 1080pembree: Pathtracer ISPC - Asian Dragondaphne: OpenMP - NDT Mappingonednn: IP Shapes 1D - f32 - CPUonednn: IP Shapes 1D - bf16bf16bf16 - CPUonednn: IP Shapes 1D - u8s8f32 - CPUtensorflow: CPU - 16 - GoogLeNetsvt-av1: Preset 8 - Bosphorus 4Ktensorflow: CPU - 32 - AlexNetonednn: IP Shapes 3D - bf16bf16bf16 - CPUonednn: IP Shapes 3D - f32 - CPUonednn: IP Shapes 3D - u8s8f32 - CPUtensorflow: CPU - 16 - AlexNetdraco: Church Facadeopencv: Videodraco: Liononednn: Convolution Batch Shapes Auto - f32 - CPUonednn: Convolution Batch Shapes Auto - u8s8f32 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUsvt-av1: Preset 8 - Bosphorus 1080psvt-av1: Preset 12 - Bosphorus 4Ksvt-av1: Preset 13 - Bosphorus 4Konednn: Deconvolution Batch shapes_3d - bf16bf16bf16 - CPUonednn: Deconvolution Batch shapes_3d - f32 - CPUonednn: Deconvolution Batch shapes_3d - u8s8f32 - CPUsvt-av1: Preset 12 - Bosphorus 1080psvt-av1: Preset 13 - Bosphorus 1080pspecfem3d: Layered Halfspaceefghij4764.872.87606.472.72403.978335.322295.654218.0910.20247.58194.87393187.577241748164.8539.48191.8939.33192.5920.64122.3441.74181.48895509542.01180.302973860.21207730858801112229882502674437821803036404443801522141681008346380060188005217.0586896.837578.044.74771.579884886712367887876.55164243.2160858.82160594.88185846.03186259.01151096.19189134.12128779.98156.11179.61430.06903.519886.91891.69959.84513.735511.49508.7462073056.279732.02394.99749522103532.7330587566.1969.444268751.7113099.4070335084317194.519.04638.99113.1644.6353.8334127000111117980169730047327810034691218106366292147448000419714.913789.3631563.049213.1913.339178.0028.3762.6313699619612132.2344291685166885.1237.924728.01412742871789.1375346.36142321.264835927.286279831.9754460.73729.26119392.91121611.7678610054.562223417.17403838439091.08969776.95492399.21620.6690.33157.05189308144344.18166.9977939.26895520.286280163.344044040396119310.974466781.04489496.0752534.2278931999.996083.67276.833501538.25113551.29774.15152594.25969913.814035573.8438963.93334310.6429.4092204.4635.003135.470625.4397.720594.538530.414775468.667.23144.40431165.731.308393.11271.59894191.5955.794393.951.683551.630060.536328301.946654840854721.849221.633515.61518127.605191.683176.7467.772041.881990.412985502.718480.1194753.672.89596.9772.78405.544336.192295.08218.210.20247.53194.12393187.047239683160.3139.45192.0339.48191.8520.57122.7641.88180.8841.85181.0173154442808783.97207833238401111710567102675039578803032828141601522077130808348167710188565217.0187396.64576.644.75571.4980385385112265687076.93164831.27160487.19160596.18185721.16186542.92151139.57190175.9129247.95160.27169.83437.85887.466897.839892.28360.17522.147510.863509.8092082786.919783.02415.33762732125283.9330347565.6769.424165587.2412910.6390120784295820.039.03640.44113.1144.6553.207411900011099948003193007132778901472197810655292114656700041983913782.4617923.049213.2113.281178.6528.2762.8413644019713932.0676292885115907.3838.096627.87212675561788.274817.2142232.064835657.846288116.3954403.04727.52119349.95121882.8678868686.92431294.874035027.5139071.55972025.68493574.131591.75712.93157.76189854144244.5163.3977970.36905447.596274228.794042140460119310.274463109.62490342.6152157.8479082149.286066.73275.583529363.29113258.03742.73152593.8969633.434027385.5938948.07334324.229.2865204.5934.953235.381925.4987.738754.312280.414756466.037.97644.40751166.691.300874.194370.831823191.9656.242393.421.647271.618190.5603923026656816254921.694781.545795.62639126.881193.997174.6497.772931.861050.408712514.742483.35310.19247.8739139.37192.4139.47191.8920.61122.5041.75181.4541.85181.012085001752011110376680026681035216030308740437015221020262083464185808668108498648872109816.639861.35464.4213009.34287640940.12112.7744.7853.1574122000110435579797030136927765704799172107867894147194000419763.713813.43.054178.4628.3013641919811432.19625204045.0738.069828.03612828691321.53722.67160.15189308144040.02184.8377787.387787657.636274788.614046040402119385.684460696.99492918.3152459.6379143065.746092.1276.133645309.63113380.23735.31152594.67969577.514023520.1239002.26334325.6429.358834.740635.50557.51544.28481142.6455.29965305525128.142193.389173.494521.86486.2525851.234.0317420.7760.7431100.0670.6136149.2154.1412.356323.7324.1012340.07889.41188.0196261.9433.8717343.9721.3110910.2253.26527.021472.555406.289137247.8622162.4411.68216.25287.69444247.78997.2949814.18202349232.0644.94168.5444.91168.6722.29113.2845.65165.9645.67165.882136467.2114535141250752523096531830351270332079941349831049889516235732289320161540159.46880107.672400.984.56652.77826809829104675890112.07268.91137681.75109256.71113684.79118060.34168568.79168427.66118589.58173426.91148306.00260.0995.60538.851264.551269.701268.1386.69705.247706.980700.3632969099.973522.26557.29675262977539.7224753386.2652.153134084.2719476.0849827013380951.409.26457.06124.5440.5567.752284633312843777104322116792353603312207072619868126293000286665.89773.234.261526893.08182.2542109.68157.2413.853202.3824.95102.1952319.0650.529375213623521.0021256565373285.4224.363334.95714987991203.5064172.35147732.483348679.103241988.2737556.08431.3081804.4584915.6623508474.852473550.012257645.9426814.161165638.38338894.541135.63961.40260.37129649163837.0393.2571159.227943561.313244081.73280132802881800.433089184.07337606.3435599.2023388075.584302.25189.502408058.8681986.25433.20104780.051165454.032257415.1826680.47334563.4719.7176154.6022.434723.069527.79311.21333.723470.494556348.918.05928.31081294.741.674604.164810.496400150.3547.308306.791.754121.860150.633800248.816390679951671.917331.794368.12585123.797173.515166.19411.29262.533420.545388540.944541.9955938.934.0417426.6560.8131134.9870.8836290.1154.1212.396343.224.0512314.63889.89188.2196362.2433.7917300.8321.3510929.6553.22533.5472.977406.189138390.8107162.4711.68216.24287.13442247.87298.6450501.6203055231.7844.97168.4444.97168.4522.28113.3445.44166.7245.62166.052164313.8314540475710752494187901828145176602081364437701050030669805730328490164124159.34864105.928400.984.56652.82822796822100766918111.89271.33138919.17109346.8113549.16118216.29173404.22173949.23118462.79182855.42150468.57273.4494.09391.511269.411271.331274.1986.62696.919709.057702.4612815652.76574.21454.13689842583399.1423632385.8852.213152408.5118470.6885564853411591.019.24552.96124.5440.5568.805285200012713867083412114762337184311133872533469125816000286768.19755.334.244549133.06182.3242145.59157.4113.886201.5325.06103.2852878.3450.649367113630820.9833245925303334.5824.281635.07614853911155.264087.97148250.983343242.273242574.7937349.03431.1981676.8385096.0123474659.322540532.652258819.8626817.831166455.13340013.71140.671013.54256.92129731163053.3594.571274.127960924.713243813.64280322800381695.623087253.76340726.635705.9823517728.164310.23190.12496277.1982002.88430.95104763.641165955.272257168.8126761.46334550.1119.7739154.9222.359623.052527.811.22823.72380.493467349.18.29628.27131247.751.670354.133380.497396150.4147.731306.641.740511.825370.645101248.156411669351761.904041.735768.14498123.112172.35166.27711.29532.527440.551504540.895537.1386087.334.117458.8560.7331093.8870.5836137.8254.1412.416354.0924.112337.84888.58187.596002.5633.8417325.7921.3910952.9653.25522.275473.739405.753139367.2041162.2511.66216.52287.41443248.55498.1550252.05200536232.1344.85168.8844.95168.5022.46112.4445.49166.5345.53166.391993539.1114546066550752749434201816474110602083449976101049950975305727203610160766159.48840106.067400.264.57452.8810810865104033903112.28269.03137744.28109558.98113739.84118198.09168958.55170495.74118453.74172379.56149032.75261.7494.02390.831269.071260.991263.7386.71702.077699.08713.0522728277.396586.72381.74677602906460.4423743386.4152.163161848.4318237.2483639673343846.789.33257.86124.9040.4368.867284300012848907065082116572337608309732271326607125180000286274.29754.134.313528783.0580.3341131.21157.3813.823200.9225.13103.1452806.1450.549359913600520.9765267084864771.1824.277935.02514926051206.8463405.19148426.953342591.63244195.8637389.38431.880822.7185213.6323468513.732570371.442256203.8626817.861165016.2341456.231137.341014.53306.3129638164013.1799.7670654.257955059.053244457.42280222801281778.723090952.6342626.1735839.5523552788.54304.94189.272386711.9981862.66431.59104750.991165479.962254827.326729.2334550.319.6193154.9922.323223.158927.9111.5773.720020.495215348.548.25828.43051246.031.677394.160950.49217150.6847.803304.841.753931.837090.624801248.496407681351731.890811.775448.13586123.271173.551167.43311.30432.535180.545626536.004537.548OpenBenchmarking.org

NWChem

NWChem is an open-source high performance computational chemistry package. Per NWChem's documentation, "NWChem aims to provide its users with computational chemistry tools that are scalable both in their ability to treat large scientific computational chemistry problems efficiently, and in their use of available parallel computing resources from high-performance parallel supercomputers to conventional workstation clusters." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.0.2Input: C240 Buckyballefhij130026003900520065004764.84753.65851.25938.96087.31. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lga -larmci -lpeigs -l64to32 -lopenblas -lpthread -lrt -llapack -lnwcblas -lmpi_usempif08 -lmpi_mpifh -lmpi -lcomex -lm -m64 -ffast-math -std=legacy -fdefault-integer-8 -finline-functions -O2

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: PLE - Data Type: BF16hij816243240SE +/- 0.03, N = 334.0334.0434.10

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: PLE - Data Type: BF16hij4K8K12K16K20KSE +/- 16.66, N = 317420.7717426.6517458.85

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: MMOE - Data Type: BF16hij1428425670SE +/- 0.10, N = 360.7460.8160.73

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: MMOE - Data Type: BF16hij7K14K21K28K35KSE +/- 51.49, N = 331100.0631134.9831093.88

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: PLE - Data Type: FP32hij1632486480SE +/- 0.14, N = 370.6170.8870.58

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: PLE - Data Type: FP32hij8K16K24K32K40KSE +/- 70.13, N = 336149.2136290.1136137.82

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: ResNet-50efhij1632486480SE +/- 0.00, N = 372.8772.8954.1454.1254.14

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: BST - Data Type: BF16hij3691215SE +/- 0.01, N = 312.3512.3912.41

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: BST - Data Type: BF16hij14002800420056007000SE +/- 2.60, N = 36323.736343.206354.09

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: DCNv2 - Data Type: BF16hij612182430SE +/- 0.02, N = 324.1024.0524.10

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: DCNv2 - Data Type: BF16hij3K6K9K12K15KSE +/- 7.04, N = 312340.0712314.6312337.84

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Barbershop - Compute: CPU-Onlyefhij2004006008001000SE +/- 4.90, N = 3606.40596.97889.41889.89888.58

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: MMOE - Data Type: FP32hij4080120160200SE +/- 0.11, N = 3188.01188.21187.50

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: MMOE - Data Type: FP32hij20K40K60K80K100KSE +/- 55.23, N = 396261.9496362.2496002.56

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: DCNv2 - Data Type: FP32hij816243240SE +/- 0.01, N = 333.8733.7933.84

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: DCNv2 - Data Type: FP32hij4K8K12K16K20KSE +/- 4.89, N = 317343.9717300.8317325.79

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

Build: allmodconfig

e: The test quit with a non-zero exit status.

f: The test quit with a non-zero exit status.

g: The test quit with a non-zero exit status.

h: The test quit with a non-zero exit status.

i: The test quit with a non-zero exit status.

j: The test quit with a non-zero exit status.

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: BST - Data Type: FP32hij510152025SE +/- 0.06, N = 321.3121.3521.39

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: BST - Data Type: FP32hij2K4K6K8K10KSE +/- 27.09, N = 310910.2210929.6510952.96

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 256 - Model: ResNet-50efhij1632486480SE +/- 0.01, N = 372.7272.7853.2653.2253.25

Timed LLVM Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix Makefilesefhij120240360480600SE +/- 3.74, N = 3403.98405.54527.02533.50522.28

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Ninjaefhij100200300400500SE +/- 0.23, N = 3335.32336.19472.56472.98473.74

Timed Node.js Compilation

This test profile times how long it takes to build/compile Node.js itself from source. Node.js is a JavaScript run-time built from the Chrome V8 JavaScript engine while itself is written in C/C++. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Node.js Compilation 19.8.1Time To Compileefhij90180270360450SE +/- 0.10, N = 3295.65295.08406.29406.19405.75

PETSc

PETSc, the Portable, Extensible Toolkit for Scientific Computation, is for the scalable (parallel) solution of scientific applications modeled by partial differential equations. This test profile runs the PETSc "make streams" benchmark and records the throughput rate when all available cores are utilized for the MPI Streams build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterPETSc 3.19Test: Streamshij30K60K90K120K150KSE +/- 515.13, N = 3137247.86138390.81139367.201. (CC) gcc options: -fPIC -O3 -O2 -lpthread -ludev -lm

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: GoogLeNetefhij50100150200250SE +/- 0.01, N = 3218.09218.20162.44162.47162.25

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Uploadefghij3691215SE +/- 0.01, N = 310.2010.2010.1911.6811.6811.661. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Uploadefghij50100150200250SE +/- 0.11, N = 3247.58247.53247.87216.25216.24216.521. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Pabellon Barcelona - Compute: CPU-Onlyefhij60120180240300SE +/- 0.12, N = 3194.87194.12287.69287.13287.41

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 8192efghij100200300400500SE +/- 1.02, N = 33933933914444424431. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl

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 4.0Time To Compileefhij50100150200250SE +/- 0.31, N = 3187.58187.05247.79247.87248.55

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: DLRM - Data Type: BF16hij20406080100SE +/- 0.19, N = 397.2998.6498.15

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: DLRM - Data Type: BF16hij11K22K33K44K55KSE +/- 96.98, N = 349814.1850501.6050252.05

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Stitchingefhij50K100K150K200K250KSE +/- 801.57, N = 32417482396832023492030552005361. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Classroom - Compute: CPU-Onlyefhij50100150200250SE +/- 0.19, N = 3164.85160.31232.06231.78232.13

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On Demandefghij1020304050SE +/- 0.02, N = 339.4839.4539.3744.9444.9744.851. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Video On Demandefghij4080120160200SE +/- 0.06, N = 3191.89192.03192.41168.54168.44168.881. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Platformefghij1020304050SE +/- 0.03, N = 339.3339.4839.4744.9144.9744.951. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Platformefghij4080120160200SE +/- 0.12, N = 3192.59191.85191.89168.67168.45168.501. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Uploadefghij510152025SE +/- 0.03, N = 320.6420.5720.6122.2922.2822.461. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Uploadefghij306090120150SE +/- 0.17, N = 3122.34122.76122.50113.28113.34112.441. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demandefghij1020304050SE +/- 0.09, N = 341.7441.8841.7545.6545.4445.491. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Video On Demandefghij4080120160200SE +/- 0.33, N = 3181.49180.88181.45165.96166.72166.531. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platformefghij1020304050SE +/- 0.07, N = 342.0141.8541.8545.6745.6245.531. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Platformefghij4080120160200SE +/- 0.25, N = 3180.30181.02181.01165.88166.05166.391. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:100efhij600K1200K1800K2400K3000KSE +/- 24953.13, N = 152973860.212808783.972136467.212164313.831993539.111. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA256efghij4000M8000M12000M16000M20000MSE +/- 3599683.66, N = 32077308588020783323840208500175201453514125014540475710145460665501. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20-Poly1305efghij20000M40000M60000M80000M100000MSE +/- 6468540.55, N = 31112229882501111710567101111037668007525230965375249418790752749434201. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-256-GCMefghij60000M120000M180000M240000M300000MSE +/- 253599393.56, N = 32674437821802675039578802668103521601830351270331828145176601816474110601. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: AES-128-GCMefghij70000M140000M210000M280000M350000MSE +/- 20712512.35, N = 33036404443803032828141603030874043702079941349832081364437702083449976101. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: ChaCha20efghij30000M60000M90000M120000M150000MSE +/- 8688544.64, N = 31522141681001522077130801522102026201049889516231050030669801049950975301. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenBenchmarking.orgbyte/s, More Is BetterOpenSSL 3.1Algorithm: SHA512efghij2000M4000M6000M8000M10000MSE +/- 1617191.98, N = 38346380060834816771083464185805732289320573032849057272036101. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Graph APIefhij40K80K120K160K200KSE +/- 164.80, N = 31880051885651615401641241607661. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 256 - Model: GoogLeNetefhij50100150200250SE +/- 0.02, N = 3217.05217.01159.46159.34159.48

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 512efghij2004006008001000SE +/- 9.84, N = 58688738668808648401. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl

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 0.15Time To Compileefhij20406080100SE +/- 1.14, N = 596.8496.64107.67105.93106.07

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 512 - Model: AlexNetefhij120240360480600SE +/- 0.19, N = 3578.04576.64400.98400.98400.26

VVenC

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 4K - Video Preset: Fastefhij1.06992.13983.20974.27965.3495SE +/- 0.001, N = 34.7474.7554.5664.5664.5741. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: ResNet-50efhij1632486480SE +/- 0.01, N = 371.5071.4952.7752.8252.80

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 4096efghij2004006008001000SE +/- 2.45, N = 37988038108268228101. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 2048efghij2004006008001000SE +/- 4.69, N = 38488538498097968101. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 1024efghij2004006008001000SE +/- 8.73, N = 38678518648298228651. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Image Processingefhij30K60K90K120K150KSE +/- 916.27, N = 31236781226561046751007661040331. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.0.1Clients: 256efghij2004006008001000SE +/- 10.01, N = 38788708878909189031. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lpcre2-8 -lcrypt -lsystemd -lz -lm -lssl -lcrypto -lpthread -ldl

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: Fishy Cat - Compute: CPU-Onlyefhij306090120150SE +/- 0.09, N = 376.5576.93112.07111.89112.28

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: DLRM - Data Type: FP32hij60120180240300SE +/- 0.29, N = 3268.91271.33269.03

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: DLRM - Data Type: FP32hij30K60K90K120K150KSE +/- 145.81, N = 3137681.75138919.17137744.28

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000efhij40K80K120K160K200KSE +/- 28.90, N = 3164243.20164831.27109256.71109346.80109558.981. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500efhij30K60K90K120K150KSE +/- 17.26, N = 3160858.82160487.19113684.79113549.16113739.841. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200efhij30K60K90K120K150KSE +/- 10.78, N = 3160594.88160596.18118060.34118216.29118198.091. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.56Concurrent Requests: 1000efhij40K80K120K160K200KSE +/- 303.73, N = 3185846.03185721.16168568.79173404.22168958.551. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.56Concurrent Requests: 500efhij40K80K120K160K200KSE +/- 306.33, N = 3186259.01186542.92168427.66173949.23170495.741. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100efhij30K60K90K120K150KSE +/- 33.69, N = 3151096.19151139.57118589.58118462.79118453.741. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Apache HTTP Server

This is a test of the Apache HTTPD web server. This Apache HTTPD web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.56Concurrent Requests: 200efhij40K80K120K160K200KSE +/- 387.27, N = 3189134.12190175.90173426.91182855.42172379.561. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenBenchmarking.orgRequests Per Second, More Is BetterApache HTTP Server 2.4.56Concurrent Requests: 100efhij30K60K90K120K150KSE +/- 69.71, N = 3128779.98129247.95148306.00150468.57149032.751. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Atomicefhij60120180240300SE +/- 2.20, N = 15156.11160.27260.09273.44261.741. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: NUMAefhij4080120160200SE +/- 1.20, N = 15179.61169.8395.6094.0994.021. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: MEMFDefhij120240360480600SE +/- 14.06, N = 15430.06437.85538.85391.51390.831. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

oneDNN

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

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Training - Data Type: u8s8f32 - Engine: CPUefhij30060090012001500SE +/- 2.92, N = 3903.52887.471264.551269.411269.07MIN: 849.4MIN: 844.57MIN: 1210.64MIN: 1219.37MIN: 1220.451. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Training - Data Type: f32 - Engine: CPUefhij30060090012001500SE +/- 2.73, N = 3886.91897.841269.701271.331260.99MIN: 844.74MIN: 851.56MIN: 1215.95MIN: 1221.68MIN: 1212.221. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Training - Data Type: bf16bf16bf16 - Engine: CPUefhij30060090012001500SE +/- 3.00, N = 3891.70892.281268.131274.191263.73MIN: 849.48MIN: 849.32MIN: 1214.46MIN: 1225.59MIN: 1212.791. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Blender

Blender is an open-source 3D creation and modeling software project. This test is of Blender's Cycles performance with various sample files. GPU computing via NVIDIA OptiX and NVIDIA CUDA is currently supported as well as HIP for AMD Radeon GPUs and Intel oneAPI for Intel Graphics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 3.5Blend File: BMW27 - Compute: CPU-Onlyefhij20406080100SE +/- 0.02, N = 359.8460.1786.6986.6286.71

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Inference - Data Type: bf16bf16bf16 - Engine: CPUefhij150300450600750SE +/- 0.94, N = 3513.74522.15705.25696.92702.08MIN: 495.91MIN: 502.11MIN: 674.1MIN: 668.26MIN: 673.181. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Inference - Data Type: f32 - Engine: CPUefhij150300450600750SE +/- 0.30, N = 3511.49510.86706.98709.06699.08MIN: 490.8MIN: 492.47MIN: 677.17MIN: 677.09MIN: 668.91. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Recurrent Neural Network Inference - Data Type: u8s8f32 - Engine: CPUefhij150300450600750SE +/- 2.34, N = 3508.75509.81700.36702.46713.05MIN: 489.62MIN: 489.85MIN: 667.01MIN: 672.81MIN: 680.591. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Futexefghij600K1200K1800K2400K3000KSE +/- 44269.26, N = 152073056.272082786.912109816.632969099.972815652.702728277.391. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Socket Activityefghij2K4K6K8K10KSE +/- 622.10, N = 159732.029783.029861.353522.266574.216586.721. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: MEMFDefghij120240360480600SE +/- 13.12, N = 15394.99415.33464.42557.29454.13381.741. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Coreefhij16K32K48K64K80KSE +/- 723.19, N = 374952762736752668984677601. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Futexefhij600K1200K1800K2400K3000KSE +/- 51354.36, N = 122103532.732125283.932977539.722583399.142906460.441. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Object Detectionefhij7K14K21K28K35KSE +/- 286.60, N = 1530587303472475323632237431. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 256 - Model: AlexNetefhij120240360480600SE +/- 0.07, N = 3566.19565.67386.26385.88386.41

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: ResNet-50efhij1530456075SE +/- 0.01, N = 369.4469.4252.1552.2152.16

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:10efhij900K1800K2700K3600K4500KSE +/- 26057.18, N = 34268751.714165587.243134084.273152408.513161848.431. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Darmstadt Automotive Parallel Heterogeneous Suite

DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous Suite 2021.11.02Backend: OpenMP - Kernel: Points2Imageefghij4K8K12K16K20KSE +/- 30.76, N = 313099.4112910.6413009.3419476.0818470.6918237.251. (CXX) g++ options: -O3 -std=c++11 -fopenmp

Memcached

Memcached is a high performance, distributed memory object caching system. This Memcached test profiles makes use of memtier_benchmark for excuting this CPU/memory-focused server benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:5efhij900K1800K2700K3600K4500KSE +/- 12477.68, N = 34317194.514295820.033380951.403411591.013343846.781. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

VVenC

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 4K - Video Preset: Fasterefhij3691215SE +/- 0.012, N = 39.0469.0369.2649.2459.3321. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: CPU Cacheefghij1326395265SE +/- 3.29, N = 1238.9940.4440.1257.0652.9657.861. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Liveefghij306090120150SE +/- 0.19, N = 3113.16113.11112.77124.54124.54124.901. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx265 - Scenario: Liveefghij1020304050SE +/- 0.06, N = 344.6344.6544.7840.5540.5540.431. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.1Build: defconfigefghij1530456075SE +/- 0.52, N = 353.8353.2153.1667.7568.8168.87

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: MD5efghij900K1800K2700K3600K4500KSE +/- 2962.73, N = 34127000411900041220002846333285200028430001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

RocksDB

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

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Random Fillefghij300K600K900K1200K1500KSE +/- 4097.59, N = 31111179110999411043551284377127138612848901. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Update Randomefghij200K400K600K800K1000KSE +/- 188.07, N = 38016978003197979707104327083417065081. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Random Fill Syncefghij60K120K180K240K300KSE +/- 58.27, N = 33004733007133013692116792114762116571. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Read Random Write Randomefghij600K1200K1800K2400K3000KSE +/- 6073.00, N = 32781003277890127765702353603233718423376081. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Read While Writingefghij1000K2000K3000K4000K5000KSE +/- 10221.65, N = 34691218472197847991723122070311133830973221. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Random Readefghij20M40M60M80M100MSE +/- 707833.96, N = 31063662921065529211078678947261986872533469713266071. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: HMAC-SHA512efghij30M60M90M120M150MSE +/- 405431.05, N = 31474480001465670001471940001262930001258160001251800001. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test profile makes use of the built-in "openssl speed" benchmarking capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgverify/s, More Is BetterOpenSSL 3.1Algorithm: RSA4096efghij90K180K270K360K450KSE +/- 136.10, N = 3419714.9419839.0419763.7286665.8286768.1286274.21. (CC) gcc options: -pthread -m64 -O3 -ldl

OpenBenchmarking.orgsign/s, More Is BetterOpenSSL 3.1Algorithm: RSA4096efghij3K6K9K12K15KSE +/- 2.48, N = 313789.313782.413813.49773.29755.39754.11. (CC) gcc options: -pthread -m64 -O3 -ldl

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 five times. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.4WAV To Opus Encodehij816243240SE +/- 0.01, N = 534.2634.2434.311. (CXX) g++ options: -O3 -fvisibility=hidden -logg -lm

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Features 2Defhij14K28K42K56K70KSE +/- 672.84, N = 363156617925268954913528781. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 4 - Input: Bosphorus 4Kefghij0.69321.38642.07962.77283.466SE +/- 0.015, N = 33.0493.0493.0543.0813.0613.0501. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: DIN - Data Type: BF16hij20406080100SE +/- 0.13, N = 382.2582.3280.33

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: DIN - Data Type: BF16hij9K18K27K36K45KSE +/- 67.58, N = 342109.6842145.5941131.21

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: GoogLeNetefhij50100150200250SE +/- 0.13, N = 3213.19213.21157.24157.41157.38

VVenC

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: Fastefhij48121620SE +/- 0.04, N = 313.3413.2813.8513.8913.821. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

FFmpeg

This is a benchmark of the FFmpeg multimedia framework. The FFmpeg test profile is making use of a modified version of vbench from Columbia University's Architecture and Design Lab (ARCADE) [http://arcade.cs.columbia.edu/vbench/] that is a benchmark for video-as-a-service workloads. The test profile offers the options of a range of vbench scenarios based on freely distributable video content and offers the options of using the x264 or x265 video encoders for transcoding. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Liveefghij4080120160200SE +/- 0.33, N = 3178.00178.65178.46202.38201.53200.921. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 6.0Encoder: libx264 - Scenario: Liveefghij714212835SE +/- 0.04, N = 328.3728.2728.3024.9525.0625.131. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl

DeepRec

OpenBenchmarking.orgGstep / sec, More Is BetterDeepRecModel: DIN - Data Type: FP32hij20406080100SE +/- 0.15, N = 3102.19103.28103.14

OpenBenchmarking.orgThroughput, More Is BetterDeepRecModel: DIN - Data Type: FP32hij11K22K33K44K55KSE +/- 79.54, N = 352319.0652878.3452806.14

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: ResNet-50efhij1428425670SE +/- 0.02, N = 362.6362.8450.5250.6450.54

7-Zip Compression

This is a test of 7-Zip compression/decompression with its integrated benchmark feature. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Decompression Ratingefghij30K60K90K120K150KSE +/- 74.28, N = 31369961364401364199375293671935991. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 22.01Test: Compression Ratingefghij40K80K120K160K200KSE +/- 426.89, N = 31961211971391981141362351363081360051. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer - Model: Asian Dragon Objefghij714212835SE +/- 0.04, N = 332.2332.0732.2021.0020.9820.98MIN: 31.25 / MAX: 33.72MIN: 31.19 / MAX: 33.77MIN: 31.14 / MAX: 33.69MIN: 20.84 / MAX: 21.33MIN: 20.9 / MAX: 21.18MIN: 20.89 / MAX: 21.22

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: DNN - Deep Neural Networkefhij6K12K18K24K30KSE +/- 264.62, N = 529168292882565624592267081. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Context Switchingefghij1.2M2.4M3.6M4.8M6MSE +/- 61475.84, N = 45166885.125115907.385204045.075373285.425303334.584864771.181. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer ISPC - Model: Asian Dragon Objefghij918273645SE +/- 0.03, N = 337.9238.1038.0724.3624.2824.28MIN: 37.43 / MAX: 40.04MIN: 37.56 / MAX: 40.12MIN: 37.55 / MAX: 40.11MIN: 23.83 / MAX: 26.05MIN: 23.75 / MAX: 25.83MIN: 23.75 / MAX: 25.58

Timed FFmpeg Compilation

This test times how long it takes to build the FFmpeg multimedia library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed FFmpeg Compilation 6.0Time To Compileefghij816243240SE +/- 0.03, N = 328.0127.8728.0434.9635.0835.03

RocksDB

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

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 8.0Test: Sequential Fillefghij300K600K900K1200K1500KSE +/- 5575.25, N = 31274287126755612828691498799148539114926051. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Zlibefhij400800120016002000SE +/- 1.64, N = 31789.131788.201203.501155.201206.841. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Forkingefhij16K32K48K64K80KSE +/- 221.26, N = 375346.3674817.2064172.3564087.9763405.191. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Pthreadefhij30K60K90K120K150KSE +/- 671.65, N = 3142321.26142232.06147732.48148250.98148426.951. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Hashefhij1000K2000K3000K4000K5000KSE +/- 1653.89, N = 34835927.284835657.843348679.103343242.273342591.601. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Semaphoresefhij1.3M2.6M3.9M5.2M6.5MSE +/- 1059.44, N = 36279831.976288116.393241988.273242574.793244195.861. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: CPU Stressefhij12K24K36K48K60KSE +/- 54.92, N = 354460.7354403.0437556.0837349.0337389.381. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: MMAPefhij160320480640800SE +/- 0.39, N = 3729.26727.52431.30431.19431.801. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Vector Mathefhij30K60K90K120K150KSE +/- 56.68, N = 3119392.91119349.9581804.4581676.8380822.711. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Matrix Mathefhij30K60K90K120K150KSE +/- 51.94, N = 3121611.76121882.8684915.6685096.0185213.631. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Mallocefhij20M40M60M80M100MSE +/- 12210.94, N = 378610054.5678868686.9023508474.8523474659.3223468513.731. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: CPU Cacheefhij600K1200K1800K2400K3000KSE +/- 35357.11, N = 32223417.172431294.872473550.012540532.652570371.441. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Pollefhij900K1800K2700K3600K4500KSE +/- 1054.76, N = 34038384.004035027.512257645.942258819.862256203.861. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Cryptoefhij8K16K24K32K40KSE +/- 2.51, N = 339091.0839071.5526814.1626817.8326817.861. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: Mutexefhij200K400K600K800K1000KSE +/- 265.04, N = 3969776.95972025.681165638.381166455.131165016.201. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.06Test: SENDFILEefhij110K220K330K440K550KSE +/- 82.32, N = 3492399.20493574.13338894.54340013.70341456.231. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -ljpeg -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Zlibefghij30060090012001500SE +/- 1.93, N = 31620.601591.751321.531135.631140.671137.341. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

Darmstadt Automotive Parallel Heterogeneous Suite

DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous Suite 2021.11.02Backend: OpenMP - Kernel: Euclidean Clusterefghij2004006008001000SE +/- 10.31, N = 5690.33712.93722.67961.401013.541014.531. (CXX) g++ options: -O3 -std=c++11 -fopenmp

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Atomicefghij70140210280350SE +/- 1.08, N = 3157.05157.76160.15260.37256.92306.301. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: WPA PSKefghij40K80K120K160K200KSE +/- 27.79, N = 31893081898541893081296491297311296381. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Pthreadefghij40K80K120K160K200KSE +/- 150.68, N = 3144344.18144244.50144040.02163837.03163053.35164013.171. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: NUMAefghij4080120160200SE +/- 0.91, N = 3166.99163.39184.8393.2594.5099.761. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Forkingefghij20K40K60K80K100KSE +/- 132.55, N = 377939.2077970.3077787.3871159.2271274.1270654.251. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: System V Message Passingefghij2M4M6M8M10MSE +/- 4165.71, N = 36895520.286905447.597787657.637943561.317960924.717955059.051. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Semaphoresefghij1.3M2.6M3.9M5.2M6.5MSE +/- 626.84, N = 36280163.346274228.796274788.613244081.733243813.643244457.421. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

John The Ripper

This is a benchmark of John The Ripper, which is a password cracker. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: bcryptefghij9K18K27K36K45KSE +/- 11.26, N = 34044040421404602801328032280221. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 2023.03.14Test: Blowfishefghij9K18K27K36K45KSE +/- 11.70, N = 34039640460404022802828003280121. (CC) gcc options: -m64 -lssl -lcrypto -fopenmp -lm -lrt -lz -ldl -lcrypt

Stress-NG

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Vector Mathefghij30K60K90K120K150KSE +/- 32.35, N = 3119310.97119310.27119385.6881800.4381695.6281778.721. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Hashefghij1000K2000K3000K4000K5000KSE +/- 126.06, N = 34466781.044463109.624460696.993089184.073087253.763090952.601. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: SENDFILEefghij110K220K330K440K550KSE +/- 71.80, N = 3489496.07490342.61492918.31337606.34340726.60342626.171. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: CPU Stressefghij11K22K33K44K55KSE +/- 35.16, N = 352534.2252157.8452459.6335599.2035705.9835839.551. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Mallocefghij20M40M60M80M100MSE +/- 109768.81, N = 378931999.9979082149.2879143065.7423388075.5823517728.1623552788.501. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Memory Copyingefghij13002600390052006500SE +/- 0.91, N = 36083.676066.736092.104302.254310.234304.941. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Glibc Qsort Data Sortingefghij60120180240300SE +/- 0.26, N = 3276.83275.58276.13189.50190.10189.271. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Glibc C String Functionsefghij800K1600K2400K3200K4000KSE +/- 3246.22, N = 33501538.253529363.293645309.632408058.862496277.192386711.991. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Matrix Mathefghij20K40K60K80K100KSE +/- 62.64, N = 3113551.29113258.03113380.2381986.2582002.8881862.661. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: MMAPefghij170340510680850SE +/- 0.70, N = 3774.15742.73735.31433.20430.95431.591. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Function Callefghij30K60K90K120K150KSE +/- 3.58, N = 3152594.25152593.80152594.67104780.05104763.64104750.991. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Mutexefghij200K400K600K800K1000KSE +/- 196.52, N = 3969913.81969633.43969577.511165454.031165955.271165479.961. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Pollefghij900K1800K2700K3600K4500KSE +/- 230.77, N = 34035573.844027385.594023520.122257415.182257168.812254827.301. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: Cryptoefghij8K16K24K32K40KSE +/- 22.35, N = 338963.9338948.0739002.2626680.4726761.4626729.201. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.15.04Test: x86_64 RdRandefghij70K140K210K280K350KSE +/- 10.75, N = 3334310.64334324.20334325.64334563.47334550.11334550.301. (CC) gcc options: -std=gnu99 -O2 -lm -lc -lcrypt -ldl -lEGL -lGLESv2 -lrt -lz -pthread

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer - Model: Crownefghij714212835SE +/- 0.05, N = 329.4129.2929.3619.7219.7719.62MIN: 28.72 / MAX: 31.05MIN: 28.61 / MAX: 30.83MIN: 28.73 / MAX: 30.61MIN: 19.49 / MAX: 20.5MIN: 19.64 / MAX: 20.73MIN: 19.47 / MAX: 20.27

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: GoogLeNetefhij4080120160200SE +/- 0.08, N = 3204.46204.59154.60154.92154.99

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer ISPC - Model: Crownefghij816243240SE +/- 0.04, N = 335.0034.9534.7422.4322.3622.32MIN: 34.13 / MAX: 36.64MIN: 34.07 / MAX: 36.73MIN: 33.94 / MAX: 36.36MIN: 21.85 / MAX: 24.32MIN: 21.84 / MAX: 24MIN: 21.84 / MAX: 24.02

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer - Model: Asian Dragonefghij816243240SE +/- 0.05, N = 335.4735.3835.5123.0723.0523.16MIN: 34.22 / MAX: 37.08MIN: 34.3 / MAX: 36.96MIN: 34.23 / MAX: 37MIN: 22.9 / MAX: 23.77MIN: 22.96 / MAX: 23.68MIN: 23.06 / MAX: 23.75

VVenC

OpenBenchmarking.orgFrames Per Second, More Is BetterVVenC 1.7Video Input: Bosphorus 1080p - Video Preset: Fasterefhij714212835SE +/- 0.04, N = 325.4425.5027.7927.8027.911. (CXX) g++ options: -O3 -flto -fno-fat-lto-objects -flto=auto

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_1d - Data Type: bf16bf16bf16 - Engine: CPUefhij3691215SE +/- 0.00893, N = 37.720597.7387511.2133011.2282011.57700MIN: 7.58MIN: 7.58MIN: 11.11MIN: 11.11MIN: 11.481. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_1d - Data Type: f32 - Engine: CPUefhij1.02122.04243.06364.08485.106SE +/- 0.00329, N = 34.538534.312283.723473.723803.72002MIN: 3.54MIN: 3.56MIN: 3.03MIN: 3.04MIN: 3.11. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_1d - Data Type: u8s8f32 - Engine: CPUefhij0.11140.22280.33420.44560.557SE +/- 0.000235, N = 30.4147750.4147560.4945560.4934670.495215MIN: 0.35MIN: 0.36MIN: 0.42MIN: 0.42MIN: 0.421. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 64 - Model: AlexNetefhij100200300400500SE +/- 0.39, N = 3468.66466.03348.91349.10348.54

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 4 - Input: Bosphorus 1080pefghij246810SE +/- 0.105, N = 37.2317.9767.5158.0598.2968.2581. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Embree

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 4.0.1Binary: Pathtracer ISPC - Model: Asian Dragonefghij1020304050SE +/- 0.07, N = 344.4044.4144.2828.3128.2728.43MIN: 43.8 / MAX: 46.79MIN: 43.88 / MAX: 46.85MIN: 43.73 / MAX: 46.57MIN: 27.61 / MAX: 30.41MIN: 27.6 / MAX: 30.18MIN: 27.79 / MAX: 30.42

Darmstadt Automotive Parallel Heterogeneous Suite

DAPHNE is the Darmstadt Automotive Parallel HeterogeNEous Benchmark Suite with OpenCL / CUDA / OpenMP test cases for these automotive benchmarks for evaluating programming models in context to vehicle autonomous driving capabilities. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTest Cases Per Minute, More Is BetterDarmstadt Automotive Parallel Heterogeneous Suite 2021.11.02Backend: OpenMP - Kernel: NDT Mappingefghij30060090012001500SE +/- 0.51, N = 31165.731166.691142.641294.741247.751246.031. (CXX) g++ options: -O3 -std=c++11 -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 1D - Data Type: f32 - Engine: CPUefhij0.37740.75481.13221.50961.887SE +/- 0.00645, N = 31.308391.300871.674601.670351.67739MIN: 1.19MIN: 1.17MIN: 1.45MIN: 1.44MIN: 1.451. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 1D - Data Type: bf16bf16bf16 - Engine: CPUefhij0.94371.88742.83113.77484.7185SE +/- 0.01380, N = 33.112704.194374.164814.133384.16095MIN: 2.91MIN: 2.98MIN: 4.03MIN: 4.03MIN: 4.031. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 1D - Data Type: u8s8f32 - Engine: CPUefhij0.35980.71961.07941.43921.799SE +/- 0.002966, N = 31.5989400.8318230.4964000.4973960.492170MIN: 0.67MIN: 0.53MIN: 0.39MIN: 0.4MIN: 0.391. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: GoogLeNetefhij4080120160200SE +/- 0.06, N = 3191.59191.96150.35150.41150.68

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 8 - Input: Bosphorus 4Kefghij1326395265SE +/- 0.12, N = 355.7956.2455.3047.3147.7347.801. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 32 - Model: AlexNetefhij90180270360450SE +/- 0.15, N = 3393.95393.42306.79306.64304.84

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 3D - Data Type: bf16bf16bf16 - Engine: CPUefhij0.39470.78941.18411.57881.9735SE +/- 0.00448, N = 31.683551.647271.754121.740511.75393MIN: 1.42MIN: 1.42MIN: 1.69MIN: 1.69MIN: 1.71. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 3D - Data Type: f32 - Engine: CPUefhij0.41850.8371.25551.6742.0925SE +/- 0.00420, N = 31.630061.618191.860151.825371.83709MIN: 1.58MIN: 1.57MIN: 1.79MIN: 1.77MIN: 1.761. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: IP Shapes 3D - Data Type: u8s8f32 - Engine: CPUefhij0.14510.29020.43530.58040.7255SE +/- 0.006251, N = 30.5363280.5603920.6338000.6451010.624801MIN: 0.48MIN: 0.51MIN: 0.57MIN: 0.59MIN: 0.571. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

TensorFlow

This is a benchmark of the TensorFlow deep learning framework using the TensorFlow reference benchmarks (tensorflow/benchmarks with tf_cnn_benchmarks.py). Note with the Phoronix Test Suite there is also pts/tensorflow-lite for benchmarking the TensorFlow Lite binaries if desired for complementary metrics. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgimages/sec, More Is BetterTensorFlow 2.12Device: CPU - Batch Size: 16 - Model: AlexNetefhij70140210280350SE +/- 0.29, N = 3301.94302.00248.81248.15248.49

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Church Facadeefghij14002800420056007000SE +/- 1.20, N = 36654665665306390641164071. (CXX) g++ options: -O3

OpenCV

This is a benchmark of the OpenCV (Computer Vision) library's built-in performance tests. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenCV 4.7Test: Videoefhij2K4K6K8K10KSE +/- 13.02, N = 3840881626799669368131. (CXX) g++ options: -fPIC -fsigned-char -pthread -fomit-frame-pointer -ffunction-sections -fdata-sections -msse -msse2 -msse3 -fvisibility=hidden -O3 -shared

Google Draco

Draco is a library developed by Google for compressing/decompressing 3D geometric meshes and point clouds. This test profile uses some Artec3D PLY models as the sample 3D model input formats for Draco compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterGoogle Draco 1.5.6Model: Lionefghij12002400360048006000SE +/- 5.78, N = 35472549255255167517651731. (CXX) g++ options: -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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Convolution Batch Shapes Auto - Data Type: f32 - Engine: CPUefhij0.43140.86281.29421.72562.157SE +/- 0.00961, N = 31.849221.694781.917331.904041.89081MIN: 1.79MIN: 1.62MIN: 1.75MIN: 1.74MIN: 1.731. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Convolution Batch Shapes Auto - Data Type: u8s8f32 - Engine: CPUefhij0.40370.80741.21111.61482.0185SE +/- 0.01503, N = 31.633511.545791.794361.735761.77544MIN: 1.57MIN: 1.48MIN: 1.69MIN: 1.64MIN: 1.71. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUefhij246810SE +/- 0.00039, N = 35.615185.626398.125858.144988.13586MIN: 5.59MIN: 5.59MIN: 8.12MIN: 8.12MIN: 8.121. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 8 - Input: Bosphorus 1080pefghij306090120150SE +/- 0.32, N = 3127.61126.88128.14123.80123.11123.271. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 4Kefghij4080120160200SE +/- 0.03, N = 3191.68194.00193.39173.52172.35173.551. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 4Kefghij4080120160200SE +/- 0.56, N = 3176.75174.65173.49166.19166.28167.431. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

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 toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_3d - Data Type: bf16bf16bf16 - Engine: CPUefhij3691215SE +/- 0.00717, N = 37.772047.7729311.2926011.2953011.30430MIN: 7.7MIN: 7.71MIN: 11.22MIN: 11.22MIN: 11.231. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_3d - Data Type: f32 - Engine: CPUefhij0.57041.14081.71122.28162.852SE +/- 0.00464, N = 31.881991.861052.533422.527442.53518MIN: 1.65MIN: 1.65MIN: 2.24MIN: 2.31MIN: 2.311. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 3.1Harness: Deconvolution Batch shapes_3d - Data Type: u8s8f32 - Engine: CPUefhij0.12410.24820.37230.49640.6205SE +/- 0.000311, N = 30.4129850.4087120.5453880.5515040.545626MIN: 0.34MIN: 0.34MIN: 0.45MIN: 0.45MIN: 0.451. (CXX) g++ options: -O3 -march=native -fopenmp -msse4.1 -fPIC -pie -ldl

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 12 - Input: Bosphorus 1080pefghij120240360480600SE +/- 4.52, N = 3502.72514.74521.86540.94540.90536.001. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.4Encoder Mode: Preset 13 - Input: Bosphorus 1080pefghij120240360480600SE +/- 2.33, N = 3480.12483.35486.25542.00537.14537.551. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stress-NG

Test: IO_uring

e: The test run did not produce a result.

f: The test run did not produce a result.

h: The test run did not produce a result.

i: The test run did not produce a result.

j: The test run did not produce a result.

Test: IO_uring

e: The test run did not produce a result.

f: The test run did not produce a result.

g: The test run did not produce a result.

h: The test run did not produce a result.

i: The test run did not produce a result.

j: The test run did not produce a result.

SPECFEM3D

simulates acoustic (fluid), elastic (solid), coupled acoustic/elastic, poroelastic or seismic wave propagation in any type of conforming mesh of hexahedra. This test profile currently relies on CPU-based execution for SPECFEM3D and using a variety of their built-in examples/models for benchmarking. Learn more via the OpenBenchmarking.org test page.

Model: Tomographic Model

e: The test quit with a non-zero exit status.

f: The test quit with a non-zero exit status.

g: The test quit with a non-zero exit status.

h: The test quit with a non-zero exit status.

i: The test quit with a non-zero exit status.

j: The test quit with a non-zero exit status.

Model: Mount St. Helens

e: The test quit with a non-zero exit status.

f: The test quit with a non-zero exit status.

g: The test quit with a non-zero exit status.

h: The test quit with a non-zero exit status.

i: The test quit with a non-zero exit status.

j: The test quit with a non-zero exit status.

Model: Homogeneous Halfspace

e: The test quit with a non-zero exit status.

f: The test quit with a non-zero exit status.

g: The test quit with a non-zero exit status.

h: The test quit with a non-zero exit status.

i: The test quit with a non-zero exit status.

j: The test quit with a non-zero exit status.

Model: Water-layered Halfspace

e: The test quit with a non-zero exit status.

f: The test quit with a non-zero exit status.

g: The test quit with a non-zero exit status.

h: The test quit with a non-zero exit status.

i: The test quit with a non-zero exit status.

j: The test quit with a non-zero exit status.

Model: Layered Halfspace

e: The test quit with a non-zero exit status.

f: The test quit with a non-zero exit status.

g: The test quit with a non-zero exit status.

h: The test quit with a non-zero exit status.

i: The test quit with a non-zero exit status.

j: The test quit with a non-zero exit status.

207 Results Shown

NWChem
DeepRec:
  PLE - BF16:
    Gstep / sec
    Throughput
  MMOE - BF16:
    Gstep / sec
    Throughput
  PLE - FP32:
    Gstep / sec
    Throughput
TensorFlow
DeepRec:
  BST - BF16:
    Gstep / sec
    Throughput
  DCNv2 - BF16:
    Gstep / sec
    Throughput
Blender
DeepRec:
  MMOE - FP32:
    Gstep / sec
    Throughput
  DCNv2 - FP32:
    Gstep / sec
    Throughput
  BST - FP32:
    Gstep / sec
    Throughput
TensorFlow
Timed LLVM Compilation:
  Unix Makefiles
  Ninja
Timed Node.js Compilation
PETSc
TensorFlow
FFmpeg:
  libx264 - Upload:
    FPS
    Seconds
Blender
MariaDB
Timed Godot Game Engine Compilation
DeepRec:
  DLRM - BF16:
    Gstep / sec
    Throughput
OpenCV
Blender
FFmpeg:
  libx264 - Video On Demand:
    FPS
    Seconds
  libx264 - Platform:
    FPS
    Seconds
  libx265 - Upload:
    FPS
    Seconds
  libx265 - Video On Demand:
    FPS
    Seconds
  libx265 - Platform:
    FPS
    Seconds
Memcached
OpenSSL:
  SHA256
  ChaCha20-Poly1305
  AES-256-GCM
  AES-128-GCM
  ChaCha20
  SHA512
OpenCV
TensorFlow
MariaDB
Build2
TensorFlow
VVenC
TensorFlow
MariaDB:
  4096
  2048
  1024
OpenCV
MariaDB
Blender
DeepRec:
  DLRM - FP32:
    Gstep / sec
    Throughput
nginx:
  1000
  500
  200
Apache HTTP Server:
  1000
  500
nginx
Apache HTTP Server:
  200
  100
Stress-NG:
  Atomic
  NUMA
  MEMFD
oneDNN:
  Recurrent Neural Network Training - u8s8f32 - CPU
  Recurrent Neural Network Training - f32 - CPU
  Recurrent Neural Network Training - bf16bf16bf16 - CPU
Blender
oneDNN:
  Recurrent Neural Network Inference - bf16bf16bf16 - CPU
  Recurrent Neural Network Inference - f32 - CPU
  Recurrent Neural Network Inference - u8s8f32 - CPU
Stress-NG:
  Futex
  Socket Activity
  MEMFD
OpenCV
Stress-NG
OpenCV
TensorFlow:
  CPU - 256 - AlexNet
  CPU - 32 - ResNet-50
Memcached
Darmstadt Automotive Parallel Heterogeneous Suite
Memcached
VVenC
Stress-NG
FFmpeg:
  libx265 - Live:
    FPS
    Seconds
Timed Linux Kernel Compilation
John The Ripper
RocksDB:
  Rand Fill
  Update Rand
  Rand Fill Sync
  Read Rand Write Rand
  Read While Writing
  Rand Read
John The Ripper
OpenSSL:
  RSA4096:
    verify/s
    sign/s
Opus Codec Encoding
OpenCV
SVT-AV1
DeepRec:
  DIN - BF16:
    Gstep / sec
    Throughput
TensorFlow
VVenC
FFmpeg:
  libx264 - Live:
    FPS
    Seconds
DeepRec:
  DIN - FP32:
    Gstep / sec
    Throughput
TensorFlow
7-Zip Compression:
  Decompression Rating
  Compression Rating
Embree
OpenCV
Stress-NG
Embree
Timed FFmpeg Compilation
RocksDB
Stress-NG:
  Zlib
  Forking
  Pthread
  Hash
  Semaphores
  CPU Stress
  MMAP
  Vector Math
  Matrix Math
  Malloc
  CPU Cache
  Poll
  Crypto
  Mutex
  SENDFILE
Stress-NG
Darmstadt Automotive Parallel Heterogeneous Suite
Stress-NG
John The Ripper
Stress-NG:
  Pthread
  NUMA
  Forking
  System V Message Passing
  Semaphores
John The Ripper:
  bcrypt
  Blowfish
Stress-NG:
  Vector Math
  Hash
  SENDFILE
  CPU Stress
  Malloc
  Memory Copying
  Glibc Qsort Data Sorting
  Glibc C String Functions
  Matrix Math
  MMAP
  Function Call
  Mutex
  Poll
  Crypto
  x86_64 RdRand
Embree
TensorFlow
Embree:
  Pathtracer ISPC - Crown
  Pathtracer - Asian Dragon
VVenC
oneDNN:
  Deconvolution Batch shapes_1d - bf16bf16bf16 - CPU
  Deconvolution Batch shapes_1d - f32 - CPU
  Deconvolution Batch shapes_1d - u8s8f32 - CPU
TensorFlow
SVT-AV1
Embree
Darmstadt Automotive Parallel Heterogeneous Suite
oneDNN:
  IP Shapes 1D - f32 - CPU
  IP Shapes 1D - bf16bf16bf16 - CPU
  IP Shapes 1D - u8s8f32 - CPU
TensorFlow
SVT-AV1
TensorFlow
oneDNN:
  IP Shapes 3D - bf16bf16bf16 - CPU
  IP Shapes 3D - f32 - CPU
  IP Shapes 3D - u8s8f32 - CPU
TensorFlow
Google Draco
OpenCV
Google Draco
oneDNN:
  Convolution Batch Shapes Auto - f32 - CPU
  Convolution Batch Shapes Auto - u8s8f32 - CPU
  Convolution Batch Shapes Auto - bf16bf16bf16 - CPU
SVT-AV1:
  Preset 8 - Bosphorus 1080p
  Preset 12 - Bosphorus 4K
  Preset 13 - Bosphorus 4K
oneDNN:
  Deconvolution Batch shapes_3d - bf16bf16bf16 - CPU
  Deconvolution Batch shapes_3d - f32 - CPU
  Deconvolution Batch shapes_3d - u8s8f32 - CPU
SVT-AV1:
  Preset 12 - Bosphorus 1080p
  Preset 13 - Bosphorus 1080p