AMD EPYC 72F3 Performance Benchmarks

AMD EPYC 72F3 Linux benchmarks by Michael Larabel for a Phoronix.com article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2104267-IB-UBUNTU21035
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:

Audio Encoding 4 Tests
AV1 3 Tests
Bioinformatics 2 Tests
C++ Boost Tests 2 Tests
Timed Code Compilation 5 Tests
C/C++ Compiler Tests 11 Tests
CPU Massive 18 Tests
Creator Workloads 20 Tests
Cryptography 3 Tests
Database Test Suite 3 Tests
Encoding 10 Tests
Fortran Tests 2 Tests
Game Development 4 Tests
HPC - High Performance Computing 16 Tests
Imaging 2 Tests
Machine Learning 7 Tests
Molecular Dynamics 5 Tests
MPI Benchmarks 2 Tests
Multi-Core 21 Tests
NVIDIA GPU Compute 3 Tests
Intel oneAPI 3 Tests
OpenMPI Tests 5 Tests
Programmer / Developer System Benchmarks 8 Tests
Python 2 Tests
Renderers 2 Tests
Scientific Computing 8 Tests
Server 5 Tests
Server CPU Tests 12 Tests
Single-Threaded 8 Tests
Speech 2 Tests
Telephony 2 Tests
Texture Compression 2 Tests
Video Encoding 6 Tests
Common Workstation Benchmarks 2 Tests

Statistics

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

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Disable Color Branding
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

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
EPYC 7F32
April 22 2021
  13 Hours, 46 Minutes
EPYC 72F3
April 22 2021
  11 Hours, 50 Minutes
Invert Hiding All Results Option
  12 Hours, 48 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


AMD EPYC 72F3 Performance BenchmarksOpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 72F3 8-Core @ 3.70GHz (8 Cores / 16 Threads)AMD EPYC 7F32 8-Core @ 3.70GHz (8 Cores / 16 Threads)Supermicro H12SSL-i v1.01 (2.0 BIOS)AMD Starship/Matisse8 x 16 GB DDR4-3200MT/s 18ASF2G72PDZ-3G2E13841GB Micron_9300_MTFDHAL3T8TDPASPEED2 x Broadcom NetXtreme BCM5720 2-port PCIeUbuntu 21.045.11.0-16-generic (x86_64)GNOME Shell 3.38.4X Server 1.20.71.0.2GCC 11.0.1 20210413ext41024x768ProcessorsMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerVulkanCompilerFile-SystemScreen ResolutionAMD EPYC 72F3 Performance BenchmarksSystem Logs- Transparent Huge Pages: madvise- --disable-multilib --enable-checking=release- EPYC 72F3: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xa001119 - EPYC 7F32: Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0x830104d - Python 3.9.4- EPYC 72F3: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: always-on RSB filling + srbds: Not affected + tsx_async_abort: Not affected - EPYC 7F32: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + srbds: Not affected + tsx_async_abort: Not affected

EPYC 72F3 vs. EPYC 7F32 ComparisonPhoronix Test SuiteBaseline+21.5%+21.5%+43%+43%+64.5%+64.5%AlexNet - CPU - 10085.8%GoogleNet - CPU - 10065%No - Inference - ResNet 50 - CPU62.5%tConvolve OpenMP - Gridding56.1%tConvolve OpenMP - Degridding51.9%tConvolve MT - Gridding50.5%SecureMark-TLS38.3%Speed 9 Realtime - Bosphorus 4K33.4%M.R - Path Tracer33.3%Z.C.133%32.7%tConvolve MT - Degridding32.2%DistinctUserID31.8%Speed 8 Realtime - Bosphorus 4K31.5%Pathtracer ISPC - Asian Dragon28.9%Pathtracer ISPC - Asian Dragon Obj27.9%M.R - SciVis27.5%Pathtracer - Asian Dragon27.4%Pathtracer - Asian Dragon Obj27.1%Kostya27.1%16 - 25026.8%16 - 25026.8%Pathtracer ISPC - Crown26.3%U.3.Z.C.126.2%vklBenchmark26.1%PartialTweets25.5%Danish Mood - CPU25%Exhaustive24.9%4 - 10000 - 2,5000,1 - 1000024.6%64 - 10000 - 2,5000,1 - 1000024.5%AES-256 - Decrypt24.4%San Miguel - Path Tracer24.2%LargeRand23.3%T.F.A.T.T23%AES-25622.9%Motorbike 30M22.8%NASNet Mobile22.5%WAV To MP322.3%L.E.H22.3%Pathtracer - Crown22%Thorough21.6%Medium20.8%8 - 25019.5%8 - 25019.5%P.B.S19.2%XFrog Forest - Path Tracer18.9%32 - 25018.3%32 - 25018.2%17.9%Bosphorus 4K17.7%XFrog Forest - SciVis17.6%Bosphorus 1080p17.3%WAV To FLAC16.9%UASTC 316.9%LuxCore Benchmark - CPU16.7%San Miguel - SciVis16.6%D.T15.8%NASA Streamlines - Path Tracer15.4%Enc Mode 0 - 1080p15.2%NASA Streamlines - SciVis15%Time To Compile14.9%U.4.Z.C.114.6%CoreMark Size 666 - I.P.S14.5%6, Lossless14.4%P.D.S14%ATPase Simulation - 327,506 Atoms13.6%16 - 256 - 5713.1%1 - Bosphorus 1080p12.6%12.5%2 - 256 - 5712.4%WAV To Opus Encode12.3%612.2%4 - 256 - 5712.2%1 - 256 - 5712%Twofish - Decrypt11.8%Mobilenet Float11.7%Zstd Compression 911.5%I.R.V11.4%R.C.a.P - CPU11.2%1011.2%Blowfish11.1%Blowfish - Decrypt11.1%CPU - MobileNet v211.1%ChaCha20Poly130511%100 - 250 - Read Only - Average Latency10.9%100 - 250 - Read Only10.9%Speed 6 Realtime - Bosphorus 4K10.8%Inception V410.5%8 - 256 - 5710.2%Fayalite-FIST10.2%ChaCha20Poly1305 - Decrypt10.1%Twofish10.1%V.Q.O - Bosphorus 1080p10.1%KASUMI9.9%SqueezeNet9.8%S.F.P.R9.8%Time To Compile9.6%100 - 100 - Read Only - Average Latency9.5%100 - 100 - Read Only9.5%12 digit9.4%100 - 250 - Read Write - Average Latency9.2%100 - 250 - Read Write9.1%tConvolve MPI - Gridding9%VMAF Optimized - Bosphorus 1080p9%100 - 50 - Read Only8.9%100 - 50 - Read Only - Average Latency8.8%DLSC - CPU8.8%Time To Compile8.7%100 - 100 - Read Write8.4%100 - 100 - Read Write - Average Latency8.4%P.S.O - Bosphorus 1080p8.3%KASUMI - Decrypt8%Rhodopsin Protein7.8%10 - Bosphorus 1080p7.7%Enc Mode 4 - 1080p7.7%CPU - SqueezeNet v1.17.7%Orange Juice - CPU7.7%CAST-2567.2%CAST-256 - Decrypt7.2%7 - Bosphorus 1080p6.7%Enc Mode 8 - 1080p6.5%32 - 2506.3%Time To Compile6.1%WAV To WavPack5.6%20k Atoms5.6%Time To Compile5.4%P.P.S5.1%Mobilenet Quant4.5%Wownero - 1M2.9%CaffeCaffePlaidMLASKAPASKAPASKAPSecureMarkAOM AV1OSPrayKTX-Software toktxNumpy BenchmarkASKAPsimdjsonAOM AV1EmbreeEmbreeOSPrayEmbreeEmbreesimdjsonHammerDB - MariaDBHammerDB - MariaDBEmbreeKTX-Software toktxOpenVKLsimdjsonLuxCoreRenderASTC EncoderInfluxDBInfluxDBBotanOSPraysimdjsonPyBenchBotanOpenFOAMTensorFlow LiteLAME MP3 EncodingCloverLeafEmbreeASTC EncoderASTC EncoderHammerDB - MariaDBHammerDB - MariaDBPHPBenchOSPrayHammerDB - MariaDBHammerDB - MariaDBKeyDBx265OSPrayx265FLAC Audio EncodingKTX-Software toktxLuxCoreRenderOSPraylibjpeg-turbo tjbenchOSPraySVT-AV1OSPrayTimed Erlang/OTP CompilationKTX-Software toktxCoremarklibavif avifencTimed HMMer SearchNAMDLiquid-DSPSVT-HEVCRNNoiseLiquid-DSPOpus Codec Encodinglibavif avifencLiquid-DSPLiquid-DSPBotanTensorFlow LiteKTX-Software toktxTensorFlow LiteLuxCoreRenderlibavif avifencBotanBotanTNNBotanPostgreSQL pgbenchPostgreSQL pgbenchAOM AV1TensorFlow LiteLiquid-DSPCP2K Molecular DynamicsBotanBotanSVT-VP9BotanTensorFlow LiteACES DGEMMTimed Wasmer CompilationPostgreSQL pgbenchPostgreSQL pgbenchHelsingPostgreSQL pgbenchPostgreSQL pgbenchASKAPSVT-VP9PostgreSQL pgbenchPostgreSQL pgbenchLuxCoreRenderTimed Godot Game Engine CompilationPostgreSQL pgbenchPostgreSQL pgbenchSVT-VP9BotanLAMMPS Molecular Dynamics SimulatorSVT-HEVCSVT-AV1TNNLuxCoreRenderBotanBotanSVT-HEVCSVT-AV1HammerDB - PostgreSQLTimed Linux Kernel CompilationWavPack Audio EncodingLAMMPS Molecular Dynamics SimulatorTimed Node.js CompilationHimeno BenchmarkTensorFlow LiteXmrigEPYC 72F3EPYC 7F32

AMD EPYC 72F3 Performance Benchmarkscaffe: AlexNet - CPU - 100caffe: GoogleNet - CPU - 100plaidml: No - Inference - ResNet 50 - CPUaskap: tConvolve OpenMP - Griddingaskap: tConvolve OpenMP - Degriddingaskap: tConvolve MT - Griddingsecuremark: SecureMark-TLSaom-av1: Speed 9 Realtime - Bosphorus 4Kospray: Magnetic Reconnection - Path Tracertoktx: Zstd Compression 19numpy: askap: tConvolve MT - Degriddingsimdjson: DistinctUserIDaom-av1: Speed 8 Realtime - Bosphorus 4Kembree: Pathtracer ISPC - Asian Dragonembree: Pathtracer ISPC - Asian Dragon Objospray: Magnetic Reconnection - SciVisembree: Pathtracer - Asian Dragonembree: Pathtracer - Asian Dragon Objsimdjson: Kostyahammerdb-mariadb: 16 - 250hammerdb-mariadb: 16 - 250embree: Pathtracer ISPC - Crowntoktx: UASTC 3 + Zstd Compression 19openvkl: vklBenchmarksimdjson: PartialTweetsluxcorerender: Danish Mood - CPUastcenc: Exhaustiveinfluxdb: 4 - 10000 - 2,5000,1 - 10000influxdb: 64 - 10000 - 2,5000,1 - 10000botan: AES-256 - Decryptospray: San Miguel - Path Tracersimdjson: LargeRandpybench: Total For Average Test Timesbotan: AES-256openfoam: Motorbike 30Mtensorflow-lite: NASNet Mobileencode-mp3: WAV To MP3cloverleaf: Lagrangian-Eulerian Hydrodynamicsembree: Pathtracer - Crownastcenc: Thoroughastcenc: Mediumhammerdb-mariadb: 8 - 250hammerdb-mariadb: 8 - 250phpbench: PHP Benchmark Suiteospray: XFrog Forest - Path Tracerhammerdb-mariadb: 32 - 250hammerdb-mariadb: 32 - 250keydb: x265: Bosphorus 4Kospray: XFrog Forest - SciVisx265: Bosphorus 1080pencode-flac: WAV To FLACtoktx: UASTC 3luxcorerender: LuxCore Benchmark - CPUospray: San Miguel - SciVistjbench: Decompression Throughputospray: NASA Streamlines - Path Tracersvt-av1: Enc Mode 0 - 1080pospray: NASA Streamlines - SciVisbuild-erlang: Time To Compiletoktx: UASTC 4 + Zstd Compression 19coremark: CoreMark Size 666 - Iterations Per Secondavifenc: 6, Losslesshmmer: Pfam Database Searchnamd: ATPase Simulation - 327,506 Atomsliquid-dsp: 16 - 256 - 57svt-hevc: 1 - Bosphorus 1080prnnoise: liquid-dsp: 2 - 256 - 57encode-opus: WAV To Opus Encodeavifenc: 6liquid-dsp: 4 - 256 - 57liquid-dsp: 1 - 256 - 57botan: Twofish - Decrypttensorflow-lite: Mobilenet Floattoktx: Zstd Compression 9tensorflow-lite: Inception ResNet V2luxcorerender: Rainbow Colors and Prism - CPUavifenc: 10botan: Blowfishbotan: Blowfish - Decrypttnn: CPU - MobileNet v2botan: ChaCha20Poly1305pgbench: 100 - 250 - Read Only - Average Latencypgbench: 100 - 250 - Read Onlyaom-av1: Speed 6 Realtime - Bosphorus 4Ktensorflow-lite: Inception V4liquid-dsp: 8 - 256 - 57cp2k: Fayalite-FISTbotan: ChaCha20Poly1305 - Decryptbotan: Twofishsvt-vp9: Visual Quality Optimized - Bosphorus 1080pbotan: KASUMItensorflow-lite: SqueezeNetmt-dgemm: Sustained Floating-Point Ratebuild-wasmer: Time To Compilepgbench: 100 - 100 - Read Only - Average Latencypgbench: 100 - 100 - Read Onlyhelsing: 12 digitpgbench: 100 - 250 - Read Write - Average Latencypgbench: 100 - 250 - Read Writeaskap: tConvolve MPI - Griddingsvt-vp9: VMAF Optimized - Bosphorus 1080ppgbench: 100 - 50 - Read Onlypgbench: 100 - 50 - Read Only - Average Latencyluxcorerender: DLSC - CPUbuild-godot: Time To Compilepgbench: 100 - 100 - Read Writepgbench: 100 - 100 - Read Write - Average Latencysvt-vp9: PSNR/SSIM Optimized - Bosphorus 1080pbotan: KASUMI - Decryptlammps: Rhodopsin Proteinsvt-hevc: 10 - Bosphorus 1080psvt-av1: Enc Mode 4 - 1080ptnn: CPU - SqueezeNet v1.1luxcorerender: Orange Juice - CPUbotan: CAST-256botan: CAST-256 - Decryptsvt-hevc: 7 - Bosphorus 1080psvt-av1: Enc Mode 8 - 1080pbuild-linux-kernel: Time To Compileencode-wavpack: WAV To WavPacklammps: 20k Atomsbuild-nodejs: Time To Compilehimeno: Poisson Pressure Solvertensorflow-lite: Mobilenet Quantxmrig: Wownero - 1Mdeepspeech: CPUhammerdb-postgresql: 32 - 250EPYC 72F3EPYC 7F32373831021949.855101.665542.455522.1428256043.86166.6717.918443.124950.264.9736.2713.773012.132610.5313.662312.59063.386332219195911.976416.6421454.331.25100.11691424062.01662367.26408.4821.181.118346334.44048.841606716.69930.3012.128012.61255.1916270521892757306481.5163915193550506884.1411.632.8158.246.99311.1911.4014.63228.2718264.200.12920117.065310.534411546.62100849.21191.0521.958065938866678.3917.8541365633337.13211.61927159333368262000369.9381259852.46824533806.433.012438.570436.947246.263838.0241.02124487014.7027268805316466671118.667824.760365.888137.4498.6951911153.29135454.4450.4112431935.25711.469218098358.57173.442330990.2151.86126.070250154.002181.3194.8446.868255.084.540243.2792.79146.678146.694121.8535.09689.36113.0616.877397.5024589.3628341372528173.765.25172656397694691686666.063269.043647.683668.4420435832.8812523.827333.943744.623.7727.5910.68879.48958.2610.72739.90782.66499271514179.481021.0001153.451.00125.09121143072.21335615.65152.8330.950.910265153.44759.961967628.19437.069.940915.33756.2722226284746906127581.2754048163740430033.359.882.3949.648.17513.0791.2012.55197.1149203.640.11217.39134.482355.812359517.25773456.283103.8132.224505250466677.4520.0801214766678.00913.03924213666760958333330.9881407052.75327336475.783.348394.664393.412273.505754.9131.13222089813.2730121534822600001232.345748.892332.346124.8589.7992099122.99861459.6460.452221815.74912.520199897665.74159.162139970.2341.71136.993230704.337167.3687.7956.374236.744.214262.0772.59136.822136.895114.1732.96694.77913.7976.512419.1374367.0884661433867941.066.47651617710OpenBenchmarking.org

Caffe

This is a benchmark of the Caffe deep learning framework and currently supports the AlexNet and Googlenet model and execution on both CPUs and NVIDIA GPUs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: AlexNet - Acceleration: CPU - Iterations: 100EPYC 72F3EPYC 7F3215K30K45K60K75KSE +/- 64.71, N = 3SE +/- 60.95, N = 337383694691. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

OpenBenchmarking.orgMilli-Seconds, Fewer Is BetterCaffe 2020-02-13Model: GoogleNet - Acceleration: CPU - Iterations: 100EPYC 72F3EPYC 7F3240K80K120K160K200KSE +/- 262.46, N = 3SE +/- 117.30, N = 31021941686661. (CXX) g++ options: -fPIC -O3 -rdynamic -lglog -lgflags -lprotobuf -lpthread -lsz -lz -ldl -lm -llmdb -lopenblas

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUEPYC 72F3EPYC 7F323691215SE +/- 0.05, N = 3SE +/- 0.03, N = 39.856.06

ASKAP

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

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - GriddingEPYC 72F3EPYC 7F3211002200330044005500SE +/- 51.82, N = 15SE +/- 47.75, N = 45101.663269.041. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve OpenMP - DegriddingEPYC 72F3EPYC 7F3212002400360048006000SE +/- 35.96, N = 15SE +/- 20.40, N = 45542.453647.681. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - GriddingEPYC 72F3EPYC 7F3212002400360048006000SE +/- 64.72, N = 3SE +/- 17.33, N = 35522.143668.441. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

SecureMark

SecureMark is an objective, standardized benchmarking framework for measuring the efficiency of cryptographic processing solutions developed by EEMBC. SecureMark-TLS is benchmarking Transport Layer Security performance with a focus on IoT/edge computing. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmarks, More Is BetterSecureMark 1.0.4Benchmark: SecureMark-TLSEPYC 72F3EPYC 7F3260K120K180K240K300KSE +/- 1383.28, N = 3SE +/- 426.85, N = 32825602043581. (CC) gcc options: -pedantic -O3

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4KEPYC 72F3EPYC 7F321020304050SE +/- 0.38, N = 12SE +/- 0.40, N = 643.8632.881. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: Path TracerEPYC 72F3EPYC 7F324080120160200SE +/- 0.00, N = 10166.67125.00MIN: 142.86 / MAX: 200MIN: 100 / MAX: 142.86

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 19EPYC 72F3EPYC 7F32612182430SE +/- 0.07, N = 3SE +/- 0.03, N = 317.9223.83

Numpy Benchmark

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

OpenBenchmarking.orgScore, More Is BetterNumpy BenchmarkEPYC 72F3EPYC 7F32100200300400500SE +/- 0.98, N = 3SE +/- 2.68, N = 3443.12333.94

ASKAP

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

OpenBenchmarking.orgMillion Grid Points Per Second, More Is BetterASKAP 1.0Test: tConvolve MT - DegriddingEPYC 72F3EPYC 7F3211002200330044005500SE +/- 34.90, N = 3SE +/- 7.69, N = 34950.263744.621. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

simdjson

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

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: DistinctUserIDEPYC 72F3EPYC 7F321.11832.23663.35494.47325.5915SE +/- 0.01, N = 3SE +/- 0.01, N = 34.973.771. (CXX) g++ options: -O3 -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4KEPYC 72F3EPYC 7F32816243240SE +/- 0.07, N = 3SE +/- 0.41, N = 336.2727.591. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian DragonEPYC 72F3EPYC 7F3248121620SE +/- 0.07, N = 3SE +/- 0.07, N = 313.7710.69MIN: 13.57 / MAX: 14.09MIN: 10.54 / MAX: 10.94

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: Asian Dragon ObjEPYC 72F3EPYC 7F323691215SE +/- 0.0297, N = 3SE +/- 0.0168, N = 312.13269.4895MIN: 11.99 / MAX: 12.42MIN: 9.42 / MAX: 9.59

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: Magnetic Reconnection - Renderer: SciVisEPYC 72F3EPYC 7F323691215SE +/- 0.00, N = 3SE +/- 0.00, N = 310.538.26MIN: 10 / MAX: 11.11MIN: 7.87 / MAX: 8.33

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian DragonEPYC 72F3EPYC 7F3248121620SE +/- 0.06, N = 3SE +/- 0.03, N = 313.6610.73MIN: 13.51 / MAX: 13.86MIN: 10.63 / MAX: 10.9

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: Asian Dragon ObjEPYC 72F3EPYC 7F323691215SE +/- 0.0176, N = 3SE +/- 0.0176, N = 312.59069.9078MIN: 12.52 / MAX: 12.74MIN: 9.84 / MAX: 10.01

simdjson

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

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: KostyaEPYC 72F3EPYC 7F320.76051.5212.28153.0423.8025SE +/- 0.01, N = 3SE +/- 0.00, N = 33.382.661. (CXX) g++ options: -O3 -pthread

HammerDB - MariaDB

This is a MariaDB MySQL database server benchmark making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNew Orders Per Minute, More Is BetterHammerDB - MariaDB 10.5.9Virtual Users: 16 - Warehouses: 250EPYC 72F3EPYC 7F3214K28K42K56K70KSE +/- 918.48, N = 3SE +/- 652.66, N = 463322499271. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - MariaDB 10.5.9Virtual Users: 16 - Warehouses: 250EPYC 72F3EPYC 7F3240K80K120K160K200KSE +/- 2750.97, N = 3SE +/- 1841.26, N = 41919591514171. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer ISPC - Model: CrownEPYC 72F3EPYC 7F323691215SE +/- 0.0435, N = 3SE +/- 0.0083, N = 311.97649.4810MIN: 11.46 / MAX: 12.26MIN: 9.4 / MAX: 9.64

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3 + Zstd Compression 19EPYC 72F3EPYC 7F32510152025SE +/- 0.04, N = 4SE +/- 0.04, N = 316.6421.00

OpenVKL

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

OpenBenchmarking.orgItems / Sec, More Is BetterOpenVKL 0.9Benchmark: vklBenchmarkEPYC 72F3EPYC 7F32306090120150SE +/- 0.33, N = 3145115MIN: 1 / MAX: 563MIN: 1 / MAX: 407

simdjson

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

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: PartialTweetsEPYC 72F3EPYC 7F320.97431.94862.92293.89724.8715SE +/- 0.00, N = 3SE +/- 0.00, N = 34.333.451. (CXX) g++ options: -O3 -pthread

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Danish Mood - Acceleration: CPUEPYC 72F3EPYC 7F320.28130.56260.84391.12521.4065SE +/- 0.02, N = 3SE +/- 0.01, N = 31.251.00MIN: 0.31 / MAX: 1.58MIN: 0.23 / MAX: 1.31

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: ExhaustiveEPYC 72F3EPYC 7F32306090120150SE +/- 0.05, N = 3SE +/- 0.02, N = 3100.12125.091. (CXX) g++ options: -O3 -flto -pthread

InfluxDB

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

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000EPYC 72F3EPYC 7F32300K600K900K1200K1500KSE +/- 4799.90, N = 3SE +/- 1375.34, N = 31424062.01143072.2

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000EPYC 72F3EPYC 7F32400K800K1200K1600K2000KSE +/- 4165.96, N = 3SE +/- 869.47, N = 31662367.21335615.6

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256 - DecryptEPYC 72F3EPYC 7F3214002800420056007000SE +/- 16.31, N = 3SE +/- 6.94, N = 36408.485152.831. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: Path TracerEPYC 72F3EPYC 7F320.26550.5310.79651.0621.3275SE +/- 0.00, N = 3SE +/- 0.00, N = 31.180.95MIN: 1.16 / MAX: 1.2MIN: 0.93 / MAX: 0.96

simdjson

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

OpenBenchmarking.orgGB/s, More Is Bettersimdjson 0.8.2Throughput Test: LargeRandomEPYC 72F3EPYC 7F320.24980.49960.74940.99921.249SE +/- 0.00, N = 3SE +/- 0.00, N = 31.110.901. (CXX) g++ options: -O3 -pthread

PyBench

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

OpenBenchmarking.orgMilliseconds, Fewer Is BetterPyBench 2018-02-16Total For Average Test TimesEPYC 72F3EPYC 7F322004006008001000SE +/- 2.40, N = 3SE +/- 2.96, N = 38341026

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: AES-256EPYC 72F3EPYC 7F3214002800420056007000SE +/- 1.43, N = 3SE +/- 3.40, N = 36334.445153.451. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OpenFOAM

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 8Input: Motorbike 30MEPYC 72F3EPYC 7F321326395265SE +/- 0.07, N = 3SE +/- 0.24, N = 348.8459.961. (CXX) g++ options: -std=c++11 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lgenericPatchFields -ldynamicMesh -lmeshTools -lOpenFOAM -ldl -lm

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: NASNet MobileEPYC 72F3EPYC 7F3240K80K120K160K200KSE +/- 103.00, N = 3SE +/- 1893.07, N = 3160671196762

LAME MP3 Encoding

LAME is an MP3 encoder licensed under the LGPL. This test measures the time required to encode a WAV file to MP3 format. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3EPYC 72F3EPYC 7F32246810SE +/- 0.009, N = 6SE +/- 0.011, N = 66.6998.1941. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lncurses -lm

CloverLeaf

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

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsEPYC 72F3EPYC 7F32918273645SE +/- 0.13, N = 3SE +/- 0.19, N = 330.3037.061. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

Embree

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

OpenBenchmarking.orgFrames Per Second, More Is BetterEmbree 3.9.0Binary: Pathtracer - Model: CrownEPYC 72F3EPYC 7F323691215SE +/- 0.0188, N = 3SE +/- 0.0330, N = 312.12809.9409MIN: 11.99 / MAX: 12.37MIN: 9.84 / MAX: 10.1

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: ThoroughEPYC 72F3EPYC 7F3248121620SE +/- 0.00, N = 4SE +/- 0.00, N = 312.6115.341. (CXX) g++ options: -O3 -flto -pthread

OpenBenchmarking.orgSeconds, Fewer Is BetterASTC Encoder 2.4Preset: MediumEPYC 72F3EPYC 7F32246810SE +/- 0.0025, N = 7SE +/- 0.0095, N = 75.19166.27221. (CXX) g++ options: -O3 -flto -pthread

HammerDB - MariaDB

This is a MariaDB MySQL database server benchmark making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - MariaDB 10.5.9Virtual Users: 8 - Warehouses: 250EPYC 72F3EPYC 7F3260K120K180K240K300KSE +/- 556.76, N = 3SE +/- 375.83, N = 32705212262841. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgNew Orders Per Minute, More Is BetterHammerDB - MariaDB 10.5.9Virtual Users: 8 - Warehouses: 250EPYC 72F3EPYC 7F3220K40K60K80K100KSE +/- 212.35, N = 3SE +/- 72.58, N = 389275746901. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

PHPBench

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

OpenBenchmarking.orgScore, More Is BetterPHPBench 0.8.1PHP Benchmark SuiteEPYC 72F3EPYC 7F32160K320K480K640K800KSE +/- 4873.12, N = 3SE +/- 2155.25, N = 3730648612758

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: Path TracerEPYC 72F3EPYC 7F320.33980.67961.01941.35921.699SE +/- 0.00, N = 3SE +/- 0.00, N = 31.511.27MIN: 1.47 / MAX: 1.53MIN: 1.24 / MAX: 1.28

HammerDB - MariaDB

This is a MariaDB MySQL database server benchmark making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNew Orders Per Minute, More Is BetterHammerDB - MariaDB 10.5.9Virtual Users: 32 - Warehouses: 250EPYC 72F3EPYC 7F3214K28K42K56K70KSE +/- 845.67, N = 9SE +/- 810.22, N = 963915540481. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - MariaDB 10.5.9Virtual Users: 32 - Warehouses: 250EPYC 72F3EPYC 7F3240K80K120K160K200KSE +/- 2632.45, N = 9SE +/- 2437.10, N = 91935501637401. (CXX) g++ options: -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

KeyDB

A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.0.16EPYC 72F3EPYC 7F32110K220K330K440K550KSE +/- 1632.83, N = 3SE +/- 4394.97, N = 3506884.14430033.351. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 4KEPYC 72F3EPYC 7F323691215SE +/- 0.13, N = 7SE +/- 0.14, N = 311.639.881. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: XFrog Forest - Renderer: SciVisEPYC 72F3EPYC 7F320.63231.26461.89692.52923.1615SE +/- 0.00, N = 3SE +/- 0.00, N = 32.812.39MIN: 2.7 / MAX: 2.85MIN: 2.3 / MAX: 2.42

x265

This is a simple test of the x265 encoder run on the CPU with 1080p and 4K options for H.265 video encode performance with x265. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is Betterx265 3.4Video Input: Bosphorus 1080pEPYC 72F3EPYC 7F321326395265SE +/- 0.30, N = 5SE +/- 0.26, N = 458.2449.641. (CXX) g++ options: -O3 -rdynamic -lpthread -lrt -ldl -lnuma

FLAC Audio Encoding

This test times how long it takes to encode a sample WAV file to FLAC format five times. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACEPYC 72F3EPYC 7F32246810SE +/- 0.005, N = 6SE +/- 0.005, N = 66.9938.1751. (CXX) g++ options: -O2 -fvisibility=hidden -logg -lm

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 3EPYC 72F3EPYC 7F323691215SE +/- 0.00, N = 5SE +/- 0.01, N = 411.1913.08

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: LuxCore Benchmark - Acceleration: CPUEPYC 72F3EPYC 7F320.3150.630.9451.261.575SE +/- 0.00, N = 3SE +/- 0.00, N = 31.401.20MIN: 0.4 / MAX: 1.73MIN: 0.32 / MAX: 1.51

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: San Miguel - Renderer: SciVisEPYC 72F3EPYC 7F3248121620SE +/- 0.07, N = 3SE +/- 0.05, N = 314.6312.55MIN: 14.08 / MAX: 15.63MIN: 12.05 / MAX: 13.33

libjpeg-turbo tjbench

tjbench is a JPEG decompression/compression benchmark that is part of libjpeg-turbo, a JPEG image codec library optimized for SIMD instructions on modern CPU architectures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 2.1.0Test: Decompression ThroughputEPYC 72F3EPYC 7F3250100150200250SE +/- 0.12, N = 3SE +/- 0.74, N = 3228.27197.111. (CC) gcc options: -O3 -rdynamic

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: Path TracerEPYC 72F3EPYC 7F320.9451.892.8353.784.725SE +/- 0.01, N = 3SE +/- 0.00, N = 34.203.64MIN: 3.97 / MAX: 4.26MIN: 3.48 / MAX: 3.69

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pEPYC 72F3EPYC 7F320.0290.0580.0870.1160.145SE +/- 0.001, N = 3SE +/- 0.001, N = 60.1290.1121. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

OSPray

Intel OSPray is a portable ray-tracing engine for high-performance, high-fidenlity scientific visualizations. OSPray builds off Intel's Embree and Intel SPMD Program Compiler (ISPC) components as part of the oneAPI rendering toolkit. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOSPray 1.8.5Demo: NASA Streamlines - Renderer: SciVisEPYC 72F3EPYC 7F32510152025SE +/- 0.09, N = 420.0017.39MIN: 19.23 / MAX: 20.83MIN: 16.39 / MAX: 17.86

Timed Erlang/OTP Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Erlang/OTP Compilation 23.2Time To CompileEPYC 72F3EPYC 7F32306090120150SE +/- 0.11, N = 3SE +/- 0.26, N = 3117.07134.48

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: UASTC 4 + Zstd Compression 19EPYC 72F3EPYC 7F3280160240320400SE +/- 0.06, N = 3SE +/- 0.15, N = 3310.53355.81

Coremark

This is a test of EEMBC CoreMark processor benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations/Sec, More Is BetterCoremark 1.0CoreMark Size 666 - Iterations Per SecondEPYC 72F3EPYC 7F3290K180K270K360K450KSE +/- 1142.43, N = 3SE +/- 253.82, N = 3411546.62359517.261. (CC) gcc options: -O2 -lrt" -lrt

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6, LosslessEPYC 72F3EPYC 7F321326395265SE +/- 0.10, N = 3SE +/- 0.21, N = 349.2156.281. (CXX) g++ options: -O3 -fPIC -lm

Timed HMMer Search

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 3.3.1Pfam Database SearchEPYC 72F3EPYC 7F3220406080100SE +/- 0.05, N = 3SE +/- 0.05, N = 391.05103.811. (CC) gcc options: -O3 -pthread -lhmmer -leasel -lm

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.14ATPase Simulation - 327,506 AtomsEPYC 72F3EPYC 7F320.50051.0011.50152.0022.5025SE +/- 0.00055, N = 3SE +/- 0.00174, N = 31.958062.22450

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 16 - Buffer Length: 256 - Filter Length: 57EPYC 72F3EPYC 7F32130M260M390M520M650MSE +/- 355027.39, N = 3SE +/- 7998346.36, N = 35938866675250466671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 1 - Input: Bosphorus 1080pEPYC 72F3EPYC 7F32246810SE +/- 0.02, N = 3SE +/- 0.00, N = 38.397.451. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

RNNoise

RNNoise is a recurrent neural network for audio noise reduction developed by Mozilla and Xiph.Org. This test profile is a single-threaded test measuring the time to denoise a sample 26 minute long 16-bit RAW audio file using this recurrent neural network noise suppression library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRNNoise 2020-06-28EPYC 72F3EPYC 7F32510152025SE +/- 0.04, N = 3SE +/- 0.09, N = 317.8520.081. (CC) gcc options: -O2 -pedantic -fvisibility=hidden

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 2 - Buffer Length: 256 - Filter Length: 57EPYC 72F3EPYC 7F3230M60M90M120M150MSE +/- 552941.63, N = 3SE +/- 158570.00, N = 31365633331214766671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Opus Codec Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterOpus Codec Encoding 1.3.1WAV To Opus EncodeEPYC 72F3EPYC 7F32246810SE +/- 0.014, N = 6SE +/- 0.018, N = 67.1328.0091. (CXX) g++ options: -fvisibility=hidden -logg -lm

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 6EPYC 72F3EPYC 7F323691215SE +/- 0.03, N = 4SE +/- 0.01, N = 411.6213.041. (CXX) g++ options: -O3 -fPIC -lm

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 4 - Buffer Length: 256 - Filter Length: 57EPYC 72F3EPYC 7F3260M120M180M240M300MSE +/- 1535625.10, N = 3SE +/- 233832.80, N = 32715933332421366671. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 1 - Buffer Length: 256 - Filter Length: 57EPYC 72F3EPYC 7F3215M30M45M60M75MSE +/- 573512.28, N = 3SE +/- 121828.20, N = 368262000609583331. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Twofish - DecryptEPYC 72F3EPYC 7F3280160240320400SE +/- 0.02, N = 3SE +/- 0.10, N = 3369.94330.991. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet FloatEPYC 72F3EPYC 7F3230K60K90K120K150KSE +/- 102.61, N = 3SE +/- 136.60, N = 3125985140705

KTX-Software toktx

This is a benchmark of The Khronos Group's KTX-Software library and tools. KTX-Software provides "toktx" for converting/creating in the KTX container format for image textures. This benchmark times how long it takes to convert to KTX 2.0 format with various settings using a reference PNG sample input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterKTX-Software toktx 4.0Settings: Zstd Compression 9EPYC 72F3EPYC 7F320.61941.23881.85822.47763.097SE +/- 0.002, N = 10SE +/- 0.003, N = 102.4682.753

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception ResNet V2EPYC 72F3EPYC 7F32600K1200K1800K2400K3000KSE +/- 645.52, N = 3SE +/- 1909.73, N = 324533802733647

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Rainbow Colors and Prism - Acceleration: CPUEPYC 72F3EPYC 7F32246810SE +/- 0.06, N = 15SE +/- 0.06, N = 156.435.78MIN: 5.68 / MAX: 6.74MIN: 5.27 / MAX: 6.18

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.9.0Encoder Speed: 10EPYC 72F3EPYC 7F320.75331.50662.25993.01323.7665SE +/- 0.006, N = 9SE +/- 0.006, N = 93.0123.3481. (CXX) g++ options: -O3 -fPIC -lm

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: BlowfishEPYC 72F3EPYC 7F32100200300400500SE +/- 0.84, N = 3SE +/- 0.13, N = 3438.57394.661. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: Blowfish - DecryptEPYC 72F3EPYC 7F3290180270360450SE +/- 0.73, N = 3SE +/- 0.02, N = 3436.95393.411. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: MobileNet v2EPYC 72F3EPYC 7F3260120180240300SE +/- 0.10, N = 3SE +/- 0.83, N = 3246.26273.51MIN: 245.29 / MAX: 254.35MIN: 266.42 / MAX: 293.991. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305EPYC 72F3EPYC 7F322004006008001000SE +/- 0.87, N = 3SE +/- 0.20, N = 3838.02754.911. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

PostgreSQL pgbench

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Only - Average LatencyEPYC 72F3EPYC 7F320.25470.50940.76411.01881.2735SE +/- 0.000, N = 3SE +/- 0.001, N = 31.0211.1321. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read OnlyEPYC 72F3EPYC 7F3250K100K150K200K250KSE +/- 102.95, N = 3SE +/- 254.01, N = 32448702208981. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.0Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4KEPYC 72F3EPYC 7F3248121620SE +/- 0.07, N = 3SE +/- 0.01, N = 314.7013.271. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Inception V4EPYC 72F3EPYC 7F32600K1200K1800K2400K3000KSE +/- 212.21, N = 3SE +/- 1720.31, N = 327268803012153

Liquid-DSP

LiquidSDR's Liquid-DSP is a software-defined radio (SDR) digital signal processing library. This test profile runs a multi-threaded benchmark of this SDR/DSP library focused on embedded platform usage. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgsamples/s, More Is BetterLiquid-DSP 2021.01.31Threads: 8 - Buffer Length: 256 - Filter Length: 57EPYC 72F3EPYC 7F32110M220M330M440M550MSE +/- 2041766.66, N = 3SE +/- 340196.02, N = 35316466674822600001. (CC) gcc options: -O3 -pthread -lm -lc -lliquid

CP2K Molecular Dynamics

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

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 8.1Input: Fayalite-FISTEPYC 72F3EPYC 7F32300600900120015001118.671232.35

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: ChaCha20Poly1305 - DecryptEPYC 72F3EPYC 7F322004006008001000SE +/- 0.26, N = 3SE +/- 3.39, N = 3824.76748.891. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: TwofishEPYC 72F3EPYC 7F3280160240320400SE +/- 0.07, N = 3SE +/- 0.03, N = 3365.89332.351. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: Visual Quality Optimized - Input: Bosphorus 1080pEPYC 72F3EPYC 7F32306090120150SE +/- 0.20, N = 8SE +/- 0.09, N = 7137.44124.851. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMIEPYC 72F3EPYC 7F3220406080100SE +/- 0.02, N = 3SE +/- 0.02, N = 398.7089.801. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: SqueezeNetEPYC 72F3EPYC 7F3240K80K120K160K200KSE +/- 62.38, N = 3SE +/- 169.86, N = 3191115209912

ACES DGEMM

This is a multi-threaded DGEMM benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterACES DGEMM 1.0Sustained Floating-Point RateEPYC 72F3EPYC 7F320.74061.48122.22182.96243.703SE +/- 0.021694, N = 3SE +/- 0.037024, N = 33.2913542.9986141. (CC) gcc options: -O3 -march=native -fopenmp

Timed Wasmer Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Wasmer Compilation 1.0.2Time To CompileEPYC 72F3EPYC 7F321326395265SE +/- 0.10, N = 3SE +/- 0.53, N = 354.4559.651. (CC) gcc options: -m64 -pie -nodefaultlibs -ldl -lgcc_s -lutil -lrt -lpthread -lm -lc

PostgreSQL pgbench

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average LatencyEPYC 72F3EPYC 7F320.10130.20260.30390.40520.5065SE +/- 0.001, N = 3SE +/- 0.000, N = 30.4110.4501. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read OnlyEPYC 72F3EPYC 7F3250K100K150K200K250KSE +/- 571.75, N = 3SE +/- 45.45, N = 32431932221811. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Helsing

Helsing is an open-source POSIX vampire number generator. This test profile measures the time it takes to generate vampire numbers between varying numbers of digits. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHelsing 1.0-betaDigit Range: 12 digitEPYC 72F3EPYC 7F321.29352.5873.88055.1746.46755.2575.7491. (CC) gcc options: -O2 -pthread -lcrypto

PostgreSQL pgbench

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

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read Write - Average LatencyEPYC 72F3EPYC 7F323691215SE +/- 0.04, N = 3SE +/- 0.16, N = 411.4712.521. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 250 - Mode: Read WriteEPYC 72F3EPYC 7F325K10K15K20K25KSE +/- 75.09, N = 3SE +/- 258.53, N = 421809199891. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

ASKAP

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

OpenBenchmarking.orgMpix/sec, More Is BetterASKAP 1.0Test: tConvolve MPI - GriddingEPYC 72F3EPYC 7F322K4K6K8K10KSE +/- 64.79, N = 5SE +/- 108.48, N = 48358.577665.741. (CXX) g++ options: -O3 -fstrict-aliasing -fopenmp

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: VMAF Optimized - Input: Bosphorus 1080pEPYC 72F3EPYC 7F324080120160200SE +/- 1.17, N = 9SE +/- 0.80, N = 8173.44159.161. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read OnlyEPYC 72F3EPYC 7F3250K100K150K200K250KSE +/- 332.66, N = 3SE +/- 427.03, N = 32330992139971. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average LatencyEPYC 72F3EPYC 7F320.05270.10540.15810.21080.2635SE +/- 0.000, N = 3SE +/- 0.000, N = 30.2150.2341. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: DLSC - Acceleration: CPUEPYC 72F3EPYC 7F320.41850.8371.25551.6742.0925SE +/- 0.01, N = 3SE +/- 0.01, N = 31.861.71MIN: 1.79 / MAX: 1.94MIN: 1.62 / MAX: 1.78

Timed Godot Game Engine Compilation

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

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 3.2.3Time To CompileEPYC 72F3EPYC 7F32306090120150SE +/- 0.03, N = 3SE +/- 0.21, N = 3126.07136.99

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read WriteEPYC 72F3EPYC 7F325K10K15K20K25KSE +/- 423.48, N = 3SE +/- 191.41, N = 325015230701. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 100 - Mode: Read Write - Average LatencyEPYC 72F3EPYC 7F320.97581.95162.92743.90324.879SE +/- 0.067, N = 3SE +/- 0.036, N = 34.0024.3371. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

SVT-VP9

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-VP9 CPU-based multi-threaded video encoder for the VP9 video format with a sample YUV input video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-VP9 0.3Tuning: PSNR/SSIM Optimized - Input: Bosphorus 1080pEPYC 72F3EPYC 7F324080120160200SE +/- 0.27, N = 9SE +/- 0.19, N = 8181.31167.361. (CC) gcc options: -O3 -fcommon -fPIE -fPIC -fvisibility=hidden -pie -rdynamic -lpthread -lrt -lm

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: KASUMI - DecryptEPYC 72F3EPYC 7F3220406080100SE +/- 0.00, N = 3SE +/- 0.01, N = 394.8487.801. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: Rhodopsin ProteinEPYC 72F3EPYC 7F32246810SE +/- 0.013, N = 9SE +/- 0.003, N = 96.8686.3741. (CXX) g++ options: -O3 -pthread -lm

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 10 - Input: Bosphorus 1080pEPYC 72F3EPYC 7F3260120180240300SE +/- 0.33, N = 10SE +/- 0.20, N = 9255.08236.741. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pEPYC 72F3EPYC 7F321.02152.0433.06454.0865.1075SE +/- 0.018, N = 3SE +/- 0.027, N = 34.5404.2141. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

TNN

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

OpenBenchmarking.orgms, Fewer Is BetterTNN 0.2.3Target: CPU - Model: SqueezeNet v1.1EPYC 72F3EPYC 7F3260120180240300SE +/- 0.06, N = 3SE +/- 0.10, N = 3243.28262.08MIN: 242.94 / MAX: 243.92MIN: 260.91 / MAX: 265.711. (CXX) g++ options: -fopenmp -pthread -fvisibility=hidden -O3 -rdynamic -ldl

LuxCoreRender

LuxCoreRender is an open-source 3D physically based renderer formerly known as LuxRender. LuxCoreRender supports CPU-based rendering as well as GPU acceleration via OpenCL, NVIDIA CUDA, and NVIDIA OptiX interfaces. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgM samples/sec, More Is BetterLuxCoreRender 2.5Scene: Orange Juice - Acceleration: CPUEPYC 72F3EPYC 7F320.62781.25561.88342.51123.139SE +/- 0.01, N = 3SE +/- 0.01, N = 32.792.59MIN: 2.73 / MAX: 2.86MIN: 2.49 / MAX: 2.61

Botan

Botan is a BSD-licensed cross-platform open-source C++ crypto library "cryptography toolkit" that supports most publicly known cryptographic algorithms. The project's stated goal is to be "the best option for cryptography in C++ by offering the tools necessary to implement a range of practical systems, such as TLS protocol, X.509 certificates, modern AEAD ciphers, PKCS#11 and TPM hardware support, password hashing, and post quantum crypto schemes." Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256EPYC 72F3EPYC 7F32306090120150SE +/- 0.09, N = 3SE +/- 0.05, N = 3146.68136.821. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

OpenBenchmarking.orgMiB/s, More Is BetterBotan 2.17.3Test: CAST-256 - DecryptEPYC 72F3EPYC 7F32306090120150SE +/- 0.08, N = 3SE +/- 0.03, N = 3146.69136.901. (CXX) g++ options: -fstack-protector -m64 -pthread -lbotan-2 -ldl -lrt

SVT-HEVC

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-HEVC CPU-based multi-threaded video encoder for the HEVC / H.265 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-HEVC 1.5.0Tuning: 7 - Input: Bosphorus 1080pEPYC 72F3EPYC 7F32306090120150SE +/- 0.19, N = 7SE +/- 0.07, N = 7121.85114.171. (CC) gcc options: -fPIE -fPIC -O3 -O2 -pie -rdynamic -lpthread -lrt

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pEPYC 72F3EPYC 7F32816243240SE +/- 0.04, N = 5SE +/- 0.04, N = 535.1032.971. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.10.20Time To CompileEPYC 72F3EPYC 7F3220406080100SE +/- 1.07, N = 3SE +/- 1.33, N = 389.3694.78

WavPack Audio Encoding

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

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.3WAV To WavPackEPYC 72F3EPYC 7F3248121620SE +/- 0.00, N = 5SE +/- 0.01, N = 513.0613.801. (CXX) g++ options: -rdynamic

LAMMPS Molecular Dynamics Simulator

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

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 29Oct2020Model: 20k AtomsEPYC 72F3EPYC 7F32246810SE +/- 0.016, N = 3SE +/- 0.010, N = 36.8776.5121. (CXX) g++ options: -O3 -pthread -lm

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 15.11Time To CompileEPYC 72F3EPYC 7F3290180270360450SE +/- 0.31, N = 3SE +/- 0.42, N = 3397.50419.14

Himeno Benchmark

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

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverEPYC 72F3EPYC 7F3210002000300040005000SE +/- 67.21, N = 15SE +/- 13.57, N = 34589.364367.091. (CC) gcc options: -O3 -mavx2

TensorFlow Lite

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

OpenBenchmarking.orgMicroseconds, Fewer Is BetterTensorFlow Lite 2020-08-23Model: Mobilenet QuantEPYC 72F3EPYC 7F3230K60K90K120K150KSE +/- 79.39, N = 3SE +/- 210.60, N = 3137252143386

Xmrig

Xmrig is an open-source cross-platform CPU/GPU miner for RandomX, KawPow, CryptoNight and AstroBWT. This test profile is setup to measure the Xmlrig CPU mining performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.12.1Variant: Wownero - Hash Count: 1MEPYC 72F3EPYC 7F322K4K6K8K10KSE +/- 5.29, N = 3SE +/- 2.36, N = 38173.77941.01. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

DeepSpeech

Mozilla DeepSpeech is a speech-to-text engine powered by TensorFlow for machine learning and derived from Baidu's Deep Speech research paper. This test profile times the speech-to-text process for a roughly three minute audio recording. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterDeepSpeech 0.6Acceleration: CPUEPYC 72F3EPYC 7F321530456075SE +/- 0.69, N = 3SE +/- 0.09, N = 365.2566.48

HammerDB - PostgreSQL

This is a benchmark of the PostgreSQL database server, making use of the HammerDB benchmarking / load testing tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Minute, More Is BetterHammerDB - PostgreSQL 13Virtual Users: 32 - Warehouses: 250EPYC 72F3EPYC 7F32140K280K420K560K700KSE +/- 8426.33, N = 4SE +/- 14222.36, N = 96563976177101. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

CPU Temperature Monitor

OpenBenchmarking.orgCelsiusCPU Temperature MonitorPhoronix Test Suite System MonitoringEPYC 72F3EPYC 7F321428425670Min: 34.5 / Avg: 51.16 / Max: 65.25Min: 31 / Avg: 55.14 / Max: 71.13

131 Results Shown

Caffe:
  AlexNet - CPU - 100
  GoogleNet - CPU - 100
PlaidML
ASKAP:
  tConvolve OpenMP - Gridding
  tConvolve OpenMP - Degridding
  tConvolve MT - Gridding
SecureMark
AOM AV1
OSPray
KTX-Software toktx
Numpy Benchmark
ASKAP
simdjson
AOM AV1
Embree:
  Pathtracer ISPC - Asian Dragon
  Pathtracer ISPC - Asian Dragon Obj
OSPray
Embree:
  Pathtracer - Asian Dragon
  Pathtracer - Asian Dragon Obj
simdjson
HammerDB - MariaDB:
  16 - 250:
    New Orders Per Minute
    Transactions Per Minute
Embree
KTX-Software toktx
OpenVKL
simdjson
LuxCoreRender
ASTC Encoder
InfluxDB:
  4 - 10000 - 2,5000,1 - 10000
  64 - 10000 - 2,5000,1 - 10000
Botan
OSPray
simdjson
PyBench
Botan
OpenFOAM
TensorFlow Lite
LAME MP3 Encoding
CloverLeaf
Embree
ASTC Encoder:
  Thorough
  Medium
HammerDB - MariaDB:
  8 - 250:
    Transactions Per Minute
    New Orders Per Minute
PHPBench
OSPray
HammerDB - MariaDB:
  32 - 250:
    New Orders Per Minute
    Transactions Per Minute
KeyDB
x265
OSPray
x265
FLAC Audio Encoding
KTX-Software toktx
LuxCoreRender
OSPray
libjpeg-turbo tjbench
OSPray
SVT-AV1
OSPray
Timed Erlang/OTP Compilation
KTX-Software toktx
Coremark
libavif avifenc
Timed HMMer Search
NAMD
Liquid-DSP
SVT-HEVC
RNNoise
Liquid-DSP
Opus Codec Encoding
libavif avifenc
Liquid-DSP:
  4 - 256 - 57
  1 - 256 - 57
Botan
TensorFlow Lite
KTX-Software toktx
TensorFlow Lite
LuxCoreRender
libavif avifenc
Botan:
  Blowfish
  Blowfish - Decrypt
TNN
Botan
PostgreSQL pgbench:
  100 - 250 - Read Only - Average Latency
  100 - 250 - Read Only
AOM AV1
TensorFlow Lite
Liquid-DSP
CP2K Molecular Dynamics
Botan:
  ChaCha20Poly1305 - Decrypt
  Twofish
SVT-VP9
Botan
TensorFlow Lite
ACES DGEMM
Timed Wasmer Compilation
PostgreSQL pgbench:
  100 - 100 - Read Only - Average Latency
  100 - 100 - Read Only
Helsing
PostgreSQL pgbench:
  100 - 250 - Read Write - Average Latency
  100 - 250 - Read Write
ASKAP
SVT-VP9
PostgreSQL pgbench:
  100 - 50 - Read Only
  100 - 50 - Read Only - Average Latency
LuxCoreRender
Timed Godot Game Engine Compilation
PostgreSQL pgbench:
  100 - 100 - Read Write
  100 - 100 - Read Write - Average Latency
SVT-VP9
Botan
LAMMPS Molecular Dynamics Simulator
SVT-HEVC
SVT-AV1
TNN
LuxCoreRender
Botan:
  CAST-256
  CAST-256 - Decrypt
SVT-HEVC
SVT-AV1
Timed Linux Kernel Compilation
WavPack Audio Encoding
LAMMPS Molecular Dynamics Simulator
Timed Node.js Compilation
Himeno Benchmark
TensorFlow Lite
Xmrig
DeepSpeech
HammerDB - PostgreSQL
CPU Temperature Monitor