AMD EPYC 7601 Linux Benchmarks, Intel Xeon Gold

AMD EPYC 7601 Linux benchmarks on Ubuntu 19.04 + Linux 5.2 by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1907151-HV-EPYCVXEON47
Jump To Table - Results

View

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

Limit displaying results to tests within:

Audio Encoding 2 Tests
AV1 2 Tests
Bioinformatics 2 Tests
C++ Boost Tests 3 Tests
Chess Test Suite 6 Tests
Timed Code Compilation 4 Tests
C/C++ Compiler Tests 27 Tests
Compression Tests 4 Tests
CPU Massive 84 Tests
Creator Workloads 23 Tests
Cryptography 4 Tests
Database Test Suite 3 Tests
Encoding 8 Tests
Fortran Tests 3 Tests
Go Language Tests 2 Tests
HPC - High Performance Computing 17 Tests
Imaging 4 Tests
Java 3 Tests
Common Kernel Benchmarks 9 Tests
Machine Learning 6 Tests
Memory Test Suite 4 Tests
Molecular Dynamics 4 Tests
MPI Benchmarks 3 Tests
Multi-Core 41 Tests
NVIDIA GPU Compute 5 Tests
OpenCL 4 Tests
OpenMPI Tests 4 Tests
Programmer / Developer System Benchmarks 6 Tests
Python 5 Tests
Raytracing 3 Tests
Renderers 9 Tests
Rust Tests 2 Tests
Scientific Computing 7 Tests
Server 10 Tests
Server CPU Tests 54 Tests
Single-Threaded 21 Tests
Video Encoding 6 Tests
Common Workstation Benchmarks 8 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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
EPYC 7601
July 10 2019
  1 Day, 1 Hour, 20 Minutes
2 x EPYC 7601
July 12 2019
  1 Day, 9 Hours, 37 Minutes
2 x Xeon Gold 6138
July 14 2019
  22 Hours, 48 Minutes
Invert Hiding All Results Option
  1 Day, 3 Hours, 15 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):


AMD EPYC 7601 Linux Benchmarks, Intel Xeon GoldProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionEPYC 76012 x EPYC 76012 x Xeon Gold 6138AMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)TYAN B8026T70AE24HR (V1.02.B10 BIOS)AMD 17h8 x 16384 MB DDR4-2666MHz Samsung280GB INTEL SSDPE21D280GAllvmpipe 126GBVE2282 x Broadcom NetXtreme BCM5720 PCIeUbuntu 19.045.2.0-050200rc7-generic (x86_64) 20190630GNOME Shell 3.32.1X Server 1.20.4modesetting 1.20.43.3 Mesa 19.0.2 (LLVM 8.0 128 bits)GCC 8.3.0ext41920x10802 x AMD EPYC 7601 32-Core (64 Cores / 128 Threads)Dell 02MJ3T (1.2.5 BIOS)16 x 32 GB DDR4-2400MT/s 36ASF4G72PZ-2G6D2280GB INTEL SSDPED1D280GA + 19 x 500GB Samsung SSD 860 + 120GB SSDSCKJB120G7Rllvmpipe 504GB2 x Broadcom BCM57416 NetXtreme-E Dual-Media 10G RDMA + 2 x Broadcom NetXtreme BCM5720 PCIe 1920x10802 x Intel Xeon Gold 6138 @ 3.70GHz (40 Cores / 80 Threads)TYAN S7106 (V2.00.B20 BIOS)Intel Sky Lake-E DMI3 Registers12 x 8192 MB DDR4-2666MT/s Micron 9ASF1G72PZ-2G6B1280GB INTEL SSDPED1D280GAllvmpipe 93GB2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE3.3 Mesa 19.0.2 (LLVM 8.0 256 bits)1920x1080OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,brig,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic --without-cuda-driver -v Disk Details- NONE / errors=remount-ro,relatime,rwProcessor Details- EPYC 7601: Scaling Governor: acpi-cpufreq ondemand- 2 x Xeon Gold 6138: Scaling Governor: intel_pstate powersaveJava Details- OpenJDK Runtime Environment (build 11.0.3+7-Ubuntu-1ubuntu219.04.1)Python Details- Python 2.7.16 + Python 3.7.3Security Details- EPYC 7601: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling- 2 x EPYC 7601: l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling- 2 x Xeon Gold 6138: l1tf: Mitigation of PTE Inversion; VMX: conditional cache flushes SMT vulnerable + mds: Mitigation of Clear buffers; SMT vulnerable + meltdown: Mitigation of PTI + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of __user pointer sanitization + spectre_v2: Mitigation of Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling

EPYC 76012 x EPYC 76012 x Xeon Gold 6138Logarithmic Result OverviewPhoronix Test Suite 10.6.1MKL-DNNctx_clockRust Prime BenchmarkCore-LatencySVT-VP9MBWHimeno BenchmarkCP2K Molecular DynamicsJohn The RipperC-RayBLAKE2m-queensNAMDMemcached mcperfOpenSSLPrimesieveCloverLeafTTSIOD 3D RendererasmFishStreamBRL-CADSmallptBlogBenchBlenderStockfishebizzyPOV-RayMariaDBCpuminer-OptChaos Group V-RAYTachyonGo BenchmarksDarktableMultichase Pointer ChaserStress-NGTimed Linux Kernel CompilationSysbenchdav1dPostgreSQL pgbench7-Zip CompressionNumpy BenchmarkPHPBenchRenaissanceTensorflowParboilTimed LLVM CompilationSVT-HEVCSPECjbb 2015GNU MPCHigh Performance Conjugate GradientTSCPSVT-AV1PyBenchx264Hierarchical INTegrationHackbenchLAME MP3 EncodingZstd CompressionTimed GCC CompilationApache Benchmarklibjpeg-turbo tjbenchNode.js Octane BenchmarkCraftySunflow Rendering SystemTesseract OCRApache SiegeTinymembenchRadiance BenchmarkNumenta Anomaly BenchmarkCppPerformanceBenchmarksGraphicsMagickRedisx265NAS Parallel BenchmarksGNU GMP GMPbenchSwetEthrglibc benchAppleseedFLAC Audio EncodingTimed PHP CompilationDaCapo BenchmarkXZ CompressionRodiniaMinionlzbenchOpenCV BenchmarkFFTWRAMspeed SMPRust MandelbrotGIMPScikit-Learn

AMD EPYC 7601 Linux Benchmarks, Intel Xeon Goldblogbench: Readblogbench: Writeramspeed: Add - Integerramspeed: Copy - Integerramspeed: Scale - Integerramspeed: Triad - Integerramspeed: Average - Integerramspeed: Add - Floating Pointramspeed: Copy - Floating Pointramspeed: Scale - Floating Pointramspeed: Triad - Floating Pointramspeed: Average - Floating Pointstream: Copystream: Scalestream: Triadstream: Addtinymembench: Standard Memcpytinymembench: Standard Memsetmbw: Memory Copy - 1024 MiBmbw: Memory Copy - 4096 MiBmbw: Memory Copy - 8192 MiBethr: TCP - Latency - 1ethr: TCP - Latency - 64ethr: HTTP - Bandwidth - 1ethr: TCP - Bandwidth - 64mpcbench: Multi-Precision Benchmarknpb: BT.Anpb: EP.Cnpb: FT.Bnpb: LU.Cnpb: SP.Aparboil: OpenMP LBMparboil: OpenMP CUTCPparboil: OpenMP Stencilparboil: OpenMP MRI Griddingcloverleaf: Lagrangian-Eulerian Hydrodynamicscp2k: Fayalite-FIST Datarodinia: OpenMP LavaMDrodinia: OpenMP CFD Solverrodinia: OpenMP Streamclusterhpcg: namd: ATPase Simulation - 327,506 Atomslzbench: Brotli 0 - Compressionlzbench: Brotli 0 - Decompressionfftw: Stock - 2D FFT Size 4096fftw: Float + SSE - 2D FFT Size 4096mafft: Multiple Sequence Alignmentblake2: gmpbench: Total Timego-benchmark: jsongo-benchmark: buildgo-benchmark: garbagedacapobench: H2dacapobench: Jythondacapobench: Tradesoapdacapobench: Tradebeansrenaissance: Scala Dottyrenaissance: Apache Spark ALSrenaissance: Apache Spark Bayesrenaissance: Savina Reactors.IOrenaissance: Apache Spark PageRankrenaissance: In-Memory Database Shootoutrenaissance: Akka Unbalanced Cobwebbed Treecrafty: Elapsed Timetscp: AI Chess Performancejohn-the-ripper: Blowfishmkl-dnn: Deconvolution Batch deconv_1d - f32mkl-dnn: Convolution Batch conv_alexnet - f32mkl-dnn: Convolution Batch conv_googlenet_v3 - f32ttsiod-renderer: Phong Rendering With Soft-Shadow Mappingsvt-av1: 1080p 8-bit YUV To AV1 Video Encodesvt-hevc: 1080p 8-bit YUV To HEVC Video Encodesvt-vp9: 1080p 8-bit YUV To VP9 Video Encodex264: H.264 Video Encodingx265: H.265 1080p Video Encodingdav1d: Summer Nature 4Kdav1d: Summer Nature 1080pencode-flac: WAV To FLACencode-mp3: WAV To MP3graphics-magick: Rotategraphics-magick: Sharpengraphics-magick: Resizinghimeno: Poisson Pressure Solvercompress-7zip: Compress Speed Teststockfish: Total Timeasmfish: 1024 Hash Memory, 26 Depthswet: Averageebizzy: build-gcc: Time To Compilebuild-linux-kernel: Time To Compilebuild-php: Time To Compilebuild-llvm: Time To Compilec-ray: Total Time - 4K, 16 Rays Per Pixelprimesieve: 1e12 Prime Number Generationrust-mandel: Time To Complete Serial/Parallel Mandelbrotrust-prime: Prime Number Test To 200,000,000compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19cython-bench: hackbench: 32 - Processm-queens: Time To Solveradiance: Serialradiance: SMP Parallelrbenchmark: tachyon: Total Timeopenssl: RSA 4096-bit Performanceglibc-bench: cosglibc-bench: singlibc-bench: sqrtglibc-bench: pthread_oncecore-latency: Average Latency Between CPU Coresmultichase: 256MB Array, 256 Byte Stridemultichase: 1GB Array, 256 Byte Stride, 4 Threadstjbench: Decompression Throughputgromacs: Water Benchmarkmysqlslap: 64mysqlslap: 128mysqlslap: 256tensorflow: Cifar10pgbench: Buffer Test - Normal Load - Read Onlypgbench: Buffer Test - Normal Load - Read Writecpp-perf-bench: Math Librarycpp-perf-bench: Rand Numberscpp-perf-bench: Function Objectsdarktable: Boat - CPU-onlydarktable: Masskrug - CPU-onlydarktable: Server Rack - CPU-onlydarktable: Server Room - CPU-onlygimp: resizegimp: rotategimp: auto-levelsgimp: unsharp-maskredis: GETredis: SETstress-ng: Cryptostress-ng: CPU Stressstress-ng: Memory Copyingstress-ng: Socket Activitystress-ng: Context Switchingctx-clock: Context Switch Timesysbench: Memorysysbench: CPUv-ray: CPUblender: BMW27 - CPU-Onlyblender: Classroom - CPU-Onlyblender: Fishy Cat - CPU-Onlyblender: Barbershop - CPU-Onlyblender: Pabellon Barcelona - CPU-Onlyappleseed: Emilyappleseed: Disney Materialappleseed: Material Testersunflow: Global Illumination + Image Synthesismcperf: Getmcperf: Setpybench: Total For Average Test Timesnumenta-nab: Time To Completionhint: FLOAThint: DOUBLEapache: Static Web Page Servingapache-siege: 200apache-siege: 250geekbench: Multi Coregeekbench: Single Corenovabench: CPUnovabench: RAMnovabench: RAMphpbench: PHP Benchmark Suitescikit-learn: tesseract-ocr: Time To OCR 7 Imagesbrl-cad: VGR Performance Metricspec-jbb2015: SPECjbb2015-Composite max-jOPSspec-jbb2015: SPECjbb2015-Composite critical-jOPSlczero: BLASnumpy: povray: Trace Timesmallpt: Global Illumination Renderer; 128 Samplesnode-octane: minion: Gracefulminion: Solitaireminion: Quasigroupopencv-bench: cpuminer-opt: m7mcpuminer-opt: deepcpuminer-opt: lbrycpuminer-opt: skeincpuminer-opt: myr-grcpuminer-opt: sha256tEPYC 76012 x EPYC 76012 x Xeon Gold 613813694303390333297.9628912.2321865.6025947.012720233711.6628817.1923875.5128421.9828993.0782399.7882483.4295158.0293847.507502.639942.5712702.9812643.8212487.1225.1525.21449.76201464.136183745.31886.134706.8315162.93349.5536.301.9310.14355.341.38886.7613.2610.6117.930.780.909833444034408.00128783.896.863949.90324827322872115062874922786165277011104189180.1920593.594995.8327949.1528570.2415230.5228682.5657306078584553491313.53307.06130.33442.4436.50183.8684.82106.6728.9424.8710.4611.9842.601941691751006.1692383579365416063497452686190011487981002.1036.9362.21194.5624.4310.4748.7915.6826.8914.2132.9650.1927.61962.67296.880.42252.424567.7762.6362.062.742.42690.3980.7290.98135.132.1963026221335.62392285.9134809.12403.931760.5920.178.465.650.142.5513.7119.2627.9933.831889305.921315813.795878.337694.833896.5318893.0614026176.771545964329.2044742.5024444101.55201.69140.15504.82263.79232.70107.25254.461.0347685.5724710.141468179.59261826548.92655020032.8816320.4324772.0224522.1962198407231314032917839588512.4642.352405314488322827143.43853838022.334.703104756.7893.39145.62110.84865.601325032187542297987.2779757771846278513258028290.7024747.5128057.3628928.6434280.5928263.6425618.9431142.9629563.02155702.96156047.00177808.54173963.868017.9011196.8712755.8712596.8912599.9422.8022.71673.50220587.876177752.481680.972805.9930678.84358.0439.351.9210.47521.241.93876.256.8813.8922.820.610.449543394004448.47133323.896.863949.40663146222029483620153626957596466729572939580.1179190.366988.1244534.6835485.0617582.8552476.6556888118555117925712.83155.2570.35359.5538.46169.39119.19144.1835.3518.936.5112.0442.571931851051006.581272491023011331138557285290915011726967793.8323.6059.42134.2612.365.2847.098.2325.3714.4333.0443.2913.61959.18290.800.43271.459201.6062.7062.062.862.42961.1987.73180.13134.3552519515649.71528141.6617722.21402.511768.7820.199.526.810.154.5713.0718.0030.6439.761914069.871323945.5411949.0415556.612289.7729608.6134894437.841544190369.5188739.324082358.1699.0782.78303.65140.83238.7869.87422.101.2758296.6029378.971476158.15261914011.26641155930.5015363.2522645.0839802912.7243.874398836601127884843836012.742.523018456.7792.30145.22120.081868.7826480645471050706321.2715282318332403929430295.0829922.1425899.9929601.5228672.2326270.8629393.9022920.2026283.6826196.66130546.00107986.30115004.44115340.766109.439225.274748.424712.084713.2522.9823.65401.54270903.338670665.291479.072772.7618232.88303.4028.391.824.48396.580.992074.5211.8210.6713.240.850.662893584864534.70157023.993.134731.50263794721421326863857979869046827276123627868.6928973.183880.9029906.3725791.2518103.1425618.8673880991184986474931.5187.3337.24681.9849.87242.17256.77110.4832.2325.0811.8710.2632.112101761902608.2614104168022563755829036290088249829791039.7332.0353.70154.4427.168.4148.134.4723.3110.9657.6228.15764.17234.412.157815.4753.2753.142.332.09308.7772.1385.09175.3592833627233.36534529.5040829.49356.521340.1315.844.984.900.112.2213.7620.1527.7833.312246578.941656603.026177.2712901.627168.7418440.4920269094.1759311962820.7753919.743894478.79164.28118.31377.94204.85193.9998.71222.490.9894265.5051191.371087142.47366058356.12830288962.1720124.2529047.267705147743131306671459815212.3033.942351906033333914564183919.604.033932755.6877.37119.16107.19853.641931378557925405631.66181077OpenBenchmarking.org

BlogBench

BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadEPYC 76012 x EPYC 76012 x Xeon Gold 6138400K800K1200K1600K2000KSE +/- 4698.10, N = 3SE +/- 10695.80, N = 4SE +/- 19272.63, N = 7136943077184618332401. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: ReadEPYC 76012 x EPYC 76012 x Xeon Gold 6138300K600K900K1200K1500KMin: 1360325 / Avg: 1369430.33 / Max: 1375992Min: 739896 / Avg: 771846 / Max: 784621Min: 1753769 / Avg: 1833240.43 / Max: 19089331. (CC) gcc options: -O2 -pthread

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: WriteEPYC 76012 x EPYC 76012 x Xeon Gold 61388K16K24K32K40KSE +/- 1348.19, N = 3SE +/- 704.37, N = 3SE +/- 491.32, N = 33390327851392941. (CC) gcc options: -O2 -pthread
OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: WriteEPYC 76012 x EPYC 76012 x Xeon Gold 61387K14K21K28K35KMin: 31226 / Avg: 33903.33 / Max: 35519Min: 26490 / Avg: 27850.67 / Max: 28847Min: 38544 / Avg: 39294.33 / Max: 402191. (CC) gcc options: -O2 -pthread

RAMspeed SMP

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Add - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 61387K14K21K28K35K33297.9632580.0030295.081. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Add - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 6138110022003300440055004162.254072.505049.181. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30K28912.2328290.7029922.141. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Copy - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 6138110022003300440055003614.033536.344987.021. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Scale - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30K21865.6024747.5125899.991. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Scale - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 613890018002700360045002733.203093.444316.671. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Triad - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30K25947.0128057.3629601.521. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Triad - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 6138110022003300440055003243.383507.174933.591. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Average - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30K27202.0028928.6428672.231. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Average - Benchmark: IntegerEPYC 76012 x EPYC 76012 x Xeon Gold 6138100020003000400050003400.253616.084778.711. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Add - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 61387K14K21K28K35K33711.6634280.5926270.861. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Add - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 613890018002700360045004213.964285.074378.481. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Copy - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30K28817.1928263.6429393.901. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Copy - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 6138110022003300440055003602.153532.964898.981. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Scale - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 61385K10K15K20K25K23875.5125618.9422920.201. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Scale - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 613880016002400320040002984.443202.373820.031. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Triad - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 61387K14K21K28K35K28421.9831142.9626283.681. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Triad - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 613890018002700360045003552.753892.874380.611. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

OpenBenchmarking.orgMB/s, More Is BetterRAMspeed SMP 3.5.0Type: Average - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30K28993.0729563.0226196.661. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMB/s Per Channel, More Is BetterRAMspeed SMP 3.5.0Performance Per Memory Channel - Type: Average - Benchmark: Floating PointEPYC 76012 x EPYC 76012 x Xeon Gold 613890018002700360045003624.133695.384366.111. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

Stream

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: CopyEPYC 76012 x EPYC 76012 x Xeon Gold 613830K60K90K120K150KSE +/- 723.97, N = 5SE +/- 1142.87, N = 25SE +/- 1099.20, N = 582399.78155702.96130546.001. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s Per Channel, More Is BetterStream 2013-01-17Performance Per Memory Channel - Type: CopyEPYC 76012 x EPYC 76012 x Xeon Gold 61385K10K15K20K25K10299.9719462.8721757.671. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: CopyEPYC 76012 x EPYC 76012 x Xeon Gold 613830K60K90K120K150KMin: 80140.5 / Avg: 82399.78 / Max: 83875.6Min: 144837.2 / Avg: 155702.96 / Max: 164510.7Min: 127928.8 / Avg: 130546 / Max: 133725.61. (CC) gcc options: -O3 -march=native -fopenmp

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: ScaleEPYC 76012 x EPYC 76012 x Xeon Gold 613830K60K90K120K150KSE +/- 764.50, N = 5SE +/- 2612.71, N = 5SE +/- 836.48, N = 582483.42156047.00107986.301. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s Per Channel, More Is BetterStream 2013-01-17Performance Per Memory Channel - Type: ScaleEPYC 76012 x EPYC 76012 x Xeon Gold 61384K8K12K16K20K10310.4319505.8817997.721. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: ScaleEPYC 76012 x EPYC 76012 x Xeon Gold 613830K60K90K120K150KMin: 80152 / Avg: 82483.42 / Max: 84091.1Min: 147576.3 / Avg: 156047 / Max: 163281.9Min: 106200 / Avg: 107986.3 / Max: 111125.81. (CC) gcc options: -O3 -march=native -fopenmp

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: TriadEPYC 76012 x EPYC 76012 x Xeon Gold 613840K80K120K160K200KSE +/- 462.01, N = 5SE +/- 2134.67, N = 5SE +/- 855.03, N = 595158.02177808.54115004.441. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s Per Channel, More Is BetterStream 2013-01-17Performance Per Memory Channel - Type: TriadEPYC 76012 x EPYC 76012 x Xeon Gold 61385K10K15K20K25K11894.7522226.0719167.411. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: TriadEPYC 76012 x EPYC 76012 x Xeon Gold 613830K60K90K120K150KMin: 93885.7 / Avg: 95158.02 / Max: 96265.9Min: 172688 / Avg: 177808.54 / Max: 183274.1Min: 113159.5 / Avg: 115004.44 / Max: 117313.61. (CC) gcc options: -O3 -march=native -fopenmp

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: AddEPYC 76012 x EPYC 76012 x Xeon Gold 613840K80K120K160K200KSE +/- 720.57, N = 5SE +/- 3354.95, N = 5SE +/- 677.66, N = 593847.50173963.86115340.761. (CC) gcc options: -O3 -march=native -fopenmp
OpenBenchmarking.orgMB/s Per Channel, More Is BetterStream 2013-01-17Performance Per Memory Channel - Type: AddEPYC 76012 x EPYC 76012 x Xeon Gold 61385K10K15K20K25K11730.9421745.4819223.461. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: AddEPYC 76012 x EPYC 76012 x Xeon Gold 613830K60K90K120K150KMin: 92471.3 / Avg: 93847.5 / Max: 95777.7Min: 163845.3 / Avg: 173963.86 / Max: 184004.4Min: 113207.9 / Avg: 115340.76 / Max: 116884.51. (CC) gcc options: -O3 -march=native -fopenmp

Tinymembench

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemcpyEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KSE +/- 1.93, N = 3SE +/- 4.28, N = 3SE +/- 18.87, N = 37502.638017.906109.431. (CC) gcc options: -O2 -lm
OpenBenchmarking.orgMB/s Per Channel, More Is BetterTinymembench 2018-05-28Performance Per Memory Channel - Standard MemcpyEPYC 76012 x EPYC 76012 x Xeon Gold 61382004006008001000937.831002.241018.241. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemcpyEPYC 76012 x EPYC 76012 x Xeon Gold 613814002800420056007000Min: 7499.3 / Avg: 7502.63 / Max: 7506Min: 8010.3 / Avg: 8017.9 / Max: 8025.1Min: 6082 / Avg: 6109.43 / Max: 6145.61. (CC) gcc options: -O2 -lm

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemsetEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KSE +/- 3.38, N = 3SE +/- 10.60, N = 3SE +/- 7.28, N = 39942.5711196.879225.271. (CC) gcc options: -O2 -lm
OpenBenchmarking.orgMB/s Per Channel, More Is BetterTinymembench 2018-05-28Performance Per Memory Channel - Standard MemsetEPYC 76012 x EPYC 76012 x Xeon Gold 6138300600900120015001242.821399.611537.551. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemsetEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KMin: 9938.7 / Avg: 9942.57 / Max: 9949.3Min: 11176.6 / Avg: 11196.87 / Max: 11212.4Min: 9213 / Avg: 9225.27 / Max: 9238.21. (CC) gcc options: -O2 -lm

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 1024 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KSE +/- 2.88, N = 3SE +/- 1.31, N = 3SE +/- 7.65, N = 312702.9812755.874748.421. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s Per Channel, More Is BetterMBW 2018-09-08Performance Per Memory Channel - Test: Memory Copy - Array Size: 1024 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 6138300600900120015001587.871594.48791.401. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 1024 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KMin: 12697.34 / Avg: 12702.98 / Max: 12706.85Min: 12753.3 / Avg: 12755.87 / Max: 12757.57Min: 4739.32 / Avg: 4748.42 / Max: 4763.611. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KSE +/- 29.63, N = 3SE +/- 3.77, N = 3SE +/- 3.16, N = 312643.8212596.894712.081. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s Per Channel, More Is BetterMBW 2018-09-08Performance Per Memory Channel - Test: Memory Copy - Array Size: 4096 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 6138300600900120015001580.481574.61785.351. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KMin: 12584.55 / Avg: 12643.82 / Max: 12673.52Min: 12589.54 / Avg: 12596.89 / Max: 12602.03Min: 4707.75 / Avg: 4712.08 / Max: 4718.231. (CC) gcc options: -O3 -march=native

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 8192 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KSE +/- 187.14, N = 3SE +/- 3.67, N = 3SE +/- 4.74, N = 312487.1212599.944713.251. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s Per Channel, More Is BetterMBW 2018-09-08Performance Per Memory Channel - Test: Memory Copy - Array Size: 8192 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 6138300600900120015001560.891574.99785.541. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 8192 MiBEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KMin: 12112.86 / Avg: 12487.12 / Max: 12676.33Min: 12592.66 / Avg: 12599.94 / Max: 12604.31Min: 4707.88 / Avg: 4713.25 / Max: 4722.71. (CC) gcc options: -O3 -march=native

Ethr

Ethr is a cross-platform Golang-written network performance measurement tool developed by Microsoft that is capable of testing multiple protocols and different measurements. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1EPYC 76012 x EPYC 76012 x Xeon Gold 6138612182430SE +/- 0.39, N = 15SE +/- 0.78, N = 15SE +/- 0.49, N = 1525.1522.8022.98MIN: 12.31 / MAX: 44.94MIN: 11.87 / MAX: 47.35MIN: 12.64 / MAX: 52.72
OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 1EPYC 76012 x EPYC 76012 x Xeon Gold 6138612182430Min: 23.54 / Avg: 25.15 / Max: 28.86Min: 19.4 / Avg: 22.8 / Max: 30.28Min: 21.1 / Avg: 22.98 / Max: 26.58

OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 64EPYC 76012 x EPYC 76012 x Xeon Gold 6138612182430SE +/- 0.29, N = 15SE +/- 0.86, N = 12SE +/- 0.59, N = 1525.2122.7123.65MIN: 12.3 / MAX: 45.55MIN: 11.86 / MAX: 46.78MIN: 12.91 / MAX: 50.6
OpenBenchmarking.orgMicroseconds, Fewer Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Latency - Threads: 64EPYC 76012 x EPYC 76012 x Xeon Gold 6138612182430Min: 23.57 / Avg: 25.21 / Max: 28.09Min: 19.54 / Avg: 22.71 / Max: 28.76Min: 21.56 / Avg: 23.65 / Max: 27.14

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1EPYC 76012 x EPYC 76012 x Xeon Gold 6138150300450600750SE +/- 1.45, N = 3SE +/- 2.99, N = 3SE +/- 3.70, N = 3449.76673.50401.54MIN: 434.43 / MAX: 476.42MIN: 659.33 / MAX: 702.98MIN: 379.52 / MAX: 435.71
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: HTTP - Test: Bandwidth - Threads: 1EPYC 76012 x EPYC 76012 x Xeon Gold 6138120240360480600Min: 448.07 / Avg: 449.76 / Max: 452.64Min: 668 / Avg: 673.5 / Max: 678.28Min: 396.29 / Avg: 401.54 / Max: 408.68

OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 64EPYC 76012 x EPYC 76012 x Xeon Gold 613860K120K180K240K300KSE +/- 4686.09, N = 15SE +/- 3980.43, N = 15SE +/- 1943.68, N = 3201464.13220587.87270903.33MIN: 150920 / MAX: 225320MIN: 177620 / MAX: 258780MIN: 249200 / MAX: 278950
OpenBenchmarking.orgMbits/sec, More Is BetterEthr 2019-01-02Server Address: localhost - Protocol: TCP - Test: Bandwidth - Threads: 64EPYC 76012 x EPYC 76012 x Xeon Gold 613850K100K150K200K250KMin: 162317.33 / Avg: 201464.13 / Max: 218425.33Min: 182659.33 / Avg: 220587.87 / Max: 240807.33Min: 268010.67 / Avg: 270903.33 / Max: 274598.67

GNU MPC

GNU MPC is a C library for the arithmetic of complex numbers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGlobal Score, More Is BetterGNU MPC 1.1.0Multi-Precision BenchmarkEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KSE +/- 3.33, N = 3SE +/- 3.33, N = 3SE +/- 5.77, N = 3618361778670-mtune=k8 -march=k8-mtune=k8 -march=k8-mtune=skylake -march=broadwell1. (CC) gcc options: -lm -O2 -pedantic -fomit-frame-pointer -m64 -MT -MD -MP -MF
OpenBenchmarking.orgGlobal Score Per Core, More Is BetterGNU MPC 1.1.0Performance Per Core - Multi-Precision BenchmarkEPYC 76012 x EPYC 76012 x Xeon Gold 613850100150200250193.2296.52216.751. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgGlobal Score Per Thread, More Is BetterGNU MPC 1.1.0Performance Per Thread - Multi-Precision BenchmarkEPYC 76012 x EPYC 76012 x Xeon Gold 61382040608010096.6148.26108.381. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgGlobal Score Per GHz, More Is BetterGNU MPC 1.1.0Performance Per Clock - Multi-Precision BenchmarkEPYC 76012 x Xeon Gold 613860012001800240030002810.452343.241. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgGlobal Score Per Channel, More Is BetterGNU MPC 1.1.0Performance Per Memory Channel - Multi-Precision BenchmarkEPYC 76012 x EPYC 76012 x Xeon Gold 613830060090012001500772.88772.131445.001. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgGlobal Score, More Is BetterGNU MPC 1.1.0Multi-Precision BenchmarkEPYC 76012 x EPYC 76012 x Xeon Gold 613815003000450060007500Min: 6180 / Avg: 6183.33 / Max: 6190Min: 6170 / Avg: 6176.67 / Max: 6180Min: 8660 / Avg: 8670 / Max: 86801. (CC) gcc options: -lm -O2 -pedantic -fomit-frame-pointer -m64 -MT -MD -MP -MF

NAS Parallel Benchmarks

NPB, NAS Parallel Benchmarks, is a benchmark developed by NASA for high-end computer systems. This test profile currently uses the MPI version of NPB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: BT.AEPYC 76012 x EPYC 76012 x Xeon Gold 6138160320480640800SE +/- 1.10, N = 3SE +/- 5.65, N = 3SE +/- 0.81, N = 3745.31752.48665.291. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgTotal Mop/s Per Core, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Core - Test / Class: BT.AEPYC 76012 x EPYC 76012 x Xeon Gold 613861218243023.2911.7616.631. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgTotal Mop/s Per Thread, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Thread - Test / Class: BT.AEPYC 76012 x EPYC 76012 x Xeon Gold 6138369121511.655.888.321. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgTotal Mop/s Per GHz, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Clock - Test / Class: BT.AEPYC 76012 x Xeon Gold 613870140210280350338.78179.811. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgTotal Mop/s Per Channel, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Memory Channel - Test / Class: BT.AEPYC 76012 x EPYC 76012 x Xeon Gold 61382040608010093.1694.06110.881. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: BT.AEPYC 76012 x EPYC 76012 x Xeon Gold 6138130260390520650Min: 744.12 / Avg: 745.31 / Max: 747.51Min: 741.17 / Avg: 752.48 / Max: 758.14Min: 663.8 / Avg: 665.29 / Max: 666.61. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: EP.CEPYC 76012 x EPYC 76012 x Xeon Gold 6138400800120016002000SE +/- 4.73, N = 3SE +/- 11.28, N = 3SE +/- 1.76, N = 3886.131680.971479.071. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgTotal Mop/s Per Core, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Core - Test / Class: EP.CEPYC 76012 x EPYC 76012 x Xeon Gold 613891827364527.6926.2736.981. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgTotal Mop/s Per Thread, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Thread - Test / Class: EP.CEPYC 76012 x EPYC 76012 x Xeon Gold 613851015202513.8513.1318.491. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgTotal Mop/s Per GHz, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Clock - Test / Class: EP.CEPYC 76012 x Xeon Gold 613890180270360450402.79399.751. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgTotal Mop/s Per Channel, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Memory Channel - Test / Class: EP.CEPYC 76012 x EPYC 76012 x Xeon Gold 613850100150200250110.77210.12246.511. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: EP.CEPYC 76012 x EPYC 76012 x Xeon Gold 613830060090012001500Min: 879.94 / Avg: 886.13 / Max: 895.43Min: 1658.88 / Avg: 1680.97 / Max: 1695.97Min: 1475.86 / Avg: 1479.07 / Max: 1481.911. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: FT.BEPYC 76012 x EPYC 76012 x Xeon Gold 613810002000300040005000SE +/- 34.56, N = 3SE +/- 5.89, N = 3SE +/- 18.67, N = 34706.832805.992772.761. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgTotal Mop/s Per Core, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Core - Test / Class: FT.BEPYC 76012 x EPYC 76012 x Xeon Gold 6138306090120150147.0943.8469.321. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgTotal Mop/s Per Thread, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Thread - Test / Class: FT.BEPYC 76012 x EPYC 76012 x Xeon Gold 6138163248648073.5421.9234.661. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgTotal Mop/s Per GHz, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Clock - Test / Class: FT.BEPYC 76012 x Xeon Gold 613850010001500200025002139.47749.391. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgTotal Mop/s Per Channel, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Memory Channel - Test / Class: FT.BEPYC 76012 x EPYC 76012 x Xeon Gold 6138130260390520650588.35350.75462.131. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: FT.BEPYC 76012 x EPYC 76012 x Xeon Gold 61388001600240032004000Min: 4637.72 / Avg: 4706.83 / Max: 4741.59Min: 2795.37 / Avg: 2805.99 / Max: 2815.71Min: 2742.24 / Avg: 2772.76 / Max: 2806.661. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: LU.CEPYC 76012 x EPYC 76012 x Xeon Gold 61387K14K21K28K35KSE +/- 32.15, N = 3SE +/- 99.41, N = 3SE +/- 94.22, N = 315162.9330678.8418232.881. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgTotal Mop/s Per Core, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Core - Test / Class: LU.CEPYC 76012 x EPYC 76012 x Xeon Gold 6138100200300400500473.84479.36455.821. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgTotal Mop/s Per Thread, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Thread - Test / Class: LU.CEPYC 76012 x EPYC 76012 x Xeon Gold 613850100150200250236.92239.68227.911. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgTotal Mop/s Per GHz, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Clock - Test / Class: LU.CEPYC 76012 x Xeon Gold 6138150030004500600075006892.244927.811. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgTotal Mop/s Per Channel, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Memory Channel - Test / Class: LU.CEPYC 76012 x EPYC 76012 x Xeon Gold 613880016002400320040001895.373834.863038.811. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: LU.CEPYC 76012 x EPYC 76012 x Xeon Gold 61385K10K15K20K25KMin: 15116.98 / Avg: 15162.93 / Max: 15224.85Min: 30511.36 / Avg: 30678.84 / Max: 30855.37Min: 18049.91 / Avg: 18232.88 / Max: 18363.381. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: SP.AEPYC 76012 x EPYC 76012 x Xeon Gold 613880160240320400SE +/- 1.17, N = 3SE +/- 3.03, N = 3SE +/- 0.93, N = 3349.55358.04303.401. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
OpenBenchmarking.orgTotal Mop/s Per Core, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Core - Test / Class: SP.AEPYC 76012 x EPYC 76012 x Xeon Gold 6138369121510.925.597.591. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgTotal Mop/s Per Thread, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Thread - Test / Class: SP.AEPYC 76012 x EPYC 76012 x Xeon Gold 61381.22852.4573.68554.9146.14255.462.803.791. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgTotal Mop/s Per GHz, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Clock - Test / Class: SP.AEPYC 76012 x Xeon Gold 61384080120160200158.8982.001. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgTotal Mop/s Per Channel, More Is BetterNAS Parallel Benchmarks 3.3.1Performance Per Memory Channel - Test / Class: SP.AEPYC 76012 x EPYC 76012 x Xeon Gold 6138112233445543.6944.7650.571. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3.1Test / Class: SP.AEPYC 76012 x EPYC 76012 x Xeon Gold 613860120180240300Min: 347.21 / Avg: 349.55 / Max: 350.77Min: 353.26 / Avg: 358.04 / Max: 363.66Min: 301.67 / Avg: 303.4 / Max: 304.871. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

Parboil

The Parboil Benchmarks from the IMPACT Research Group at University of Illinois are a set of throughput computing applications for looking at computing architecture and compilers. Parboil test-cases support OpenMP, OpenCL, and CUDA multi-processing environments. However, at this time the test profile is just making use of the OpenMP and OpenCL test workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMEPYC 76012 x EPYC 76012 x Xeon Gold 6138918273645SE +/- 0.40, N = 12SE +/- 0.75, N = 15SE +/- 0.22, N = 336.3039.3528.391. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterParboil 2.5Performance Per Core - Test: OpenMP LBMEPYC 76012 x EPYC 76012 x Xeon Gold 613850010001500200025001161.62518.41135.61. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterParboil 2.5Performance Per Thread - Test: OpenMP LBMEPYC 76012 x EPYC 76012 x Xeon Gold 6138110022003300440055002323.25036.82271.21. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterParboil 2.5Performance Per Clock - Test: OpenMP LBMEPYC 76012 x Xeon Gold 61382040608010079.86105.041. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterParboil 2.5Performance Per Memory Channel - Test: OpenMP LBMEPYC 76012 x EPYC 76012 x Xeon Gold 613870140210280350290.40314.80170.341. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP LBMEPYC 76012 x EPYC 76012 x Xeon Gold 6138816243240Min: 33.22 / Avg: 36.3 / Max: 39Min: 34.29 / Avg: 39.35 / Max: 44.32Min: 27.98 / Avg: 28.39 / Max: 28.721. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPEPYC 76012 x EPYC 76012 x Xeon Gold 61380.43430.86861.30291.73722.1715SE +/- 0.03, N = 4SE +/- 0.12, N = 15SE +/- 0.02, N = 151.931.921.821. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterParboil 2.5Performance Per Core - Test: OpenMP CUTCPEPYC 76012 x EPYC 76012 x Xeon Gold 613830609012015061.76122.8872.801. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterParboil 2.5Performance Per Thread - Test: OpenMP CUTCPEPYC 76012 x EPYC 76012 x Xeon Gold 613850100150200250123.52245.76145.601. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterParboil 2.5Performance Per Clock - Test: OpenMP CUTCPEPYC 76012 x Xeon Gold 61382468104.2466.7341. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterParboil 2.5Performance Per Memory Channel - Test: OpenMP CUTCPEPYC 76012 x EPYC 76012 x Xeon Gold 61384812162015.4415.3610.921. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP CUTCPEPYC 76012 x EPYC 76012 x Xeon Gold 6138246810Min: 1.88 / Avg: 1.93 / Max: 1.98Min: 1.5 / Avg: 1.92 / Max: 2.99Min: 1.66 / Avg: 1.82 / Max: 1.931. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilEPYC 76012 x EPYC 76012 x Xeon Gold 61383691215SE +/- 0.47, N = 15SE +/- 0.37, N = 15SE +/- 0.11, N = 1510.1410.474.481. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterParboil 2.5Performance Per Core - Test: OpenMP StencilEPYC 76012 x EPYC 76012 x Xeon Gold 6138140280420560700324.48670.08179.201. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterParboil 2.5Performance Per Thread - Test: OpenMP StencilEPYC 76012 x EPYC 76012 x Xeon Gold 613830060090012001500648.961340.16358.401. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterParboil 2.5Performance Per Clock - Test: OpenMP StencilEPYC 76012 x Xeon Gold 613851015202522.3116.581. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterParboil 2.5Performance Per Memory Channel - Test: OpenMP StencilEPYC 76012 x EPYC 76012 x Xeon Gold 61382040608010081.1283.7626.881. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP StencilEPYC 76012 x EPYC 76012 x Xeon Gold 61383691215Min: 7.5 / Avg: 10.14 / Max: 14.08Min: 8.47 / Avg: 10.47 / Max: 12.75Min: 3.81 / Avg: 4.48 / Max: 5.141. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingEPYC 76012 x EPYC 76012 x Xeon Gold 6138110220330440550SE +/- 2.53, N = 3SE +/- 2.36, N = 3SE +/- 3.45, N = 3355.34521.24396.581. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterParboil 2.5Performance Per Core - Test: OpenMP MRI GriddingEPYC 76012 x EPYC 76012 x Xeon Gold 61387K14K21K28K35K11370.8833359.3615863.201. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterParboil 2.5Performance Per Thread - Test: OpenMP MRI GriddingEPYC 76012 x EPYC 76012 x Xeon Gold 613814K28K42K56K70K22741.7666718.7231726.401. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterParboil 2.5Performance Per Clock - Test: OpenMP MRI GriddingEPYC 76012 x Xeon Gold 613830060090012001500781.751467.351. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterParboil 2.5Performance Per Memory Channel - Test: OpenMP MRI GriddingEPYC 76012 x EPYC 76012 x Xeon Gold 613890018002700360045002842.724169.922379.481. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: OpenMP MRI GriddingEPYC 76012 x EPYC 76012 x Xeon Gold 613890180270360450Min: 350.46 / Avg: 355.34 / Max: 358.91Min: 516.84 / Avg: 521.24 / Max: 524.91Min: 389.71 / Avg: 396.58 / Max: 400.651. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp

CloverLeaf

CloverLeaf is a Lagrangian-Eulerian hydrodynamics benchmark. This test profile currently makes use of CloverLeaf's OpenMP version and benchmarked with the clover_bm8192.in input file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsEPYC 76012 x EPYC 76012 x Xeon Gold 61380.43430.86861.30291.73722.1715SE +/- 0.06, N = 15SE +/- 0.04, N = 15SE +/- 0.04, N = 151.381.930.991. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterCloverLeafPerformance Per Core - Lagrangian-Eulerian HydrodynamicsEPYC 76012 x EPYC 76012 x Xeon Gold 613830609012015044.16123.5239.601. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterCloverLeafPerformance Per Thread - Lagrangian-Eulerian HydrodynamicsEPYC 76012 x EPYC 76012 x Xeon Gold 61385010015020025088.32247.0479.201. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterCloverLeafPerformance Per Clock - Lagrangian-Eulerian HydrodynamicsEPYC 76012 x Xeon Gold 61380.82421.64842.47263.29684.1213.0363.6631. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterCloverLeafPerformance Per Memory Channel - Lagrangian-Eulerian HydrodynamicsEPYC 76012 x EPYC 76012 x Xeon Gold 61384812162011.0415.445.941. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterCloverLeafLagrangian-Eulerian HydrodynamicsEPYC 76012 x EPYC 76012 x Xeon Gold 6138246810Min: 0.97 / Avg: 1.38 / Max: 1.81Min: 1.71 / Avg: 1.93 / Max: 2.25Min: 0.71 / Avg: 0.99 / Max: 1.21. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp

CP2K Molecular Dynamics

CP2K is an open-source molecular dynamics software package focused on quantum chemistry and solid-state physics. This test profile currently makes use of the OpenMP implementation and using the Fayalite-FIST molecular dynamics run and measures the total time to complete. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 6.1Fayalite-FIST DataEPYC 76012 x EPYC 76012 x Xeon Gold 6138400800120016002000886.76876.252074.52
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterCP2K Molecular Dynamics 6.1Performance Per Core - Fayalite-FIST DataEPYC 76012 x EPYC 76012 x Xeon Gold 613820K40K60K80K100K28376.3256080.0082980.801. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterCP2K Molecular Dynamics 6.1Performance Per Thread - Fayalite-FIST DataEPYC 76012 x EPYC 76012 x Xeon Gold 613840K80K120K160K200K56752.64112160.00165961.601. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterCP2K Molecular Dynamics 6.1Performance Per Clock - Fayalite-FIST DataEPYC 76012 x Xeon Gold 6138160032004800640080001950.877675.721. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterCP2K Molecular Dynamics 6.1Performance Per Memory Channel - Fayalite-FIST DataEPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15K7094.087010.0012447.121. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

Rodinia

Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDEPYC 76012 x EPYC 76012 x Xeon Gold 61383691215SE +/- 0.19, N = 3SE +/- 0.12, N = 3SE +/- 0.16, N = 313.266.8811.821. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterRodinia 2.4Performance Per Core - Test: OpenMP LavaMDEPYC 76012 x EPYC 76012 x Xeon Gold 6138100200300400500424.32440.32472.801. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterRodinia 2.4Performance Per Thread - Test: OpenMP LavaMDEPYC 76012 x EPYC 76012 x Xeon Gold 61382004006008001000848.64880.64945.601. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterRodinia 2.4Performance Per Clock - Test: OpenMP LavaMDEPYC 76012 x Xeon Gold 6138102030405029.1743.731. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterRodinia 2.4Performance Per Memory Channel - Test: OpenMP LavaMDEPYC 76012 x EPYC 76012 x Xeon Gold 613820406080100106.0855.0470.921. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LavaMDEPYC 76012 x EPYC 76012 x Xeon Gold 613848121620Min: 13 / Avg: 13.26 / Max: 13.63Min: 6.65 / Avg: 6.88 / Max: 7Min: 11.61 / Avg: 11.82 / Max: 12.141. (CXX) g++ options: -O2 -lOpenCL

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD SolverEPYC 76012 x EPYC 76012 x Xeon Gold 613848121620SE +/- 0.07, N = 3SE +/- 0.51, N = 12SE +/- 0.23, N = 1510.6113.8910.671. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterRodinia 2.4Performance Per Core - Test: OpenMP CFD SolverEPYC 76012 x EPYC 76012 x Xeon Gold 61382004006008001000339.52888.96426.801. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterRodinia 2.4Performance Per Thread - Test: OpenMP CFD SolverEPYC 76012 x EPYC 76012 x Xeon Gold 6138400800120016002000679.041777.92853.601. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterRodinia 2.4Performance Per Clock - Test: OpenMP CFD SolverEPYC 76012 x Xeon Gold 613891827364523.3439.481. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterRodinia 2.4Performance Per Memory Channel - Test: OpenMP CFD SolverEPYC 76012 x EPYC 76012 x Xeon Gold 61382040608010084.88111.1264.021. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD SolverEPYC 76012 x EPYC 76012 x Xeon Gold 613848121620Min: 10.48 / Avg: 10.61 / Max: 10.72Min: 11.68 / Avg: 13.89 / Max: 18.26Min: 9.73 / Avg: 10.67 / Max: 12.361. (CXX) g++ options: -O2 -lOpenCL

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterEPYC 76012 x EPYC 76012 x Xeon Gold 6138510152025SE +/- 0.58, N = 12SE +/- 0.67, N = 15SE +/- 0.13, N = 1517.9322.8213.241. (CXX) g++ options: -O2 -lOpenCL
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterRodinia 2.4Performance Per Core - Test: OpenMP StreamclusterEPYC 76012 x EPYC 76012 x Xeon Gold 613830060090012001500573.761460.48529.601. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterRodinia 2.4Performance Per Thread - Test: OpenMP StreamclusterEPYC 76012 x EPYC 76012 x Xeon Gold 613860012001800240030001147.522920.961059.201. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterRodinia 2.4Performance Per Clock - Test: OpenMP StreamclusterEPYC 76012 x Xeon Gold 6138112233445539.4548.991. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterRodinia 2.4Performance Per Memory Channel - Test: OpenMP StreamclusterEPYC 76012 x EPYC 76012 x Xeon Gold 61384080120160200143.44182.5679.441. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterEPYC 76012 x EPYC 76012 x Xeon Gold 6138510152025Min: 15.78 / Avg: 17.93 / Max: 22.12Min: 19.88 / Avg: 22.82 / Max: 28.65Min: 12.5 / Avg: 13.24 / Max: 14.371. (CXX) g++ options: -O2 -lOpenCL

High Performance Conjugate Gradient

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

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.0EPYC 76012 x EPYC 76012 x Xeon Gold 61380.19130.38260.57390.76520.9565SE +/- 0.01, N = 15SE +/- 0.02, N = 15SE +/- 0.02, N = 150.780.610.85
OpenBenchmarking.orgGFLOP/s Per Core, More Is BetterHigh Performance Conjugate Gradient 3.0Performance Per CoreEPYC 76012 x EPYC 76012 x Xeon Gold 61380.00550.0110.01650.0220.02750.02440.00950.02131. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgGFLOP/s Per Thread, More Is BetterHigh Performance Conjugate Gradient 3.0Performance Per ThreadEPYC 76012 x EPYC 76012 x Xeon Gold 61380.00270.00540.00810.01080.01350.01220.00480.01061. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgGFLOP/s Per GHz, More Is BetterHigh Performance Conjugate Gradient 3.0Performance Per ClockEPYC 76012 x Xeon Gold 61380.07980.15960.23940.31920.3990.35450.22971. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgGFLOP/s Per Channel, More Is BetterHigh Performance Conjugate Gradient 3.0Performance Per Memory ChannelEPYC 76012 x EPYC 76012 x Xeon Gold 61380.03190.06380.09570.12760.15950.09750.07630.14171. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.0EPYC 76012 x EPYC 76012 x Xeon Gold 6138246810Min: 0.7 / Avg: 0.78 / Max: 0.85Min: 0.5 / Avg: 0.61 / Max: 0.77Min: 0.69 / Avg: 0.85 / Max: 0.96

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 2.13b1ATPase Simulation - 327,506 AtomsEPYC 76012 x EPYC 76012 x Xeon Gold 61380.20470.40940.61410.81881.0235SE +/- 0.00196, N = 3SE +/- 0.00037, N = 15SE +/- 0.00141, N = 80.909830.449540.66289
OpenBenchmarking.orgdays/ns x Core, Fewer Is BetterNAMD 2.13b1Performance Per Core - ATPase Simulation - 327,506 AtomsEPYC 76012 x EPYC 76012 x Xeon Gold 613871421283529.1228.7726.521. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgdays/ns x Thread, Fewer Is BetterNAMD 2.13b1Performance Per Thread - ATPase Simulation - 327,506 AtomsEPYC 76012 x EPYC 76012 x Xeon Gold 6138132639526558.2357.5453.031. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgdays/ns x GHz, Fewer Is BetterNAMD 2.13b1Performance Per Clock - ATPase Simulation - 327,506 AtomsEPYC 76012 x Xeon Gold 61380.55191.10381.65572.20762.75952.0022.4531. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgdays/ns x Channel, Fewer Is BetterNAMD 2.13b1Performance Per Memory Channel - ATPase Simulation - 327,506 AtomsEPYC 76012 x EPYC 76012 x Xeon Gold 61382468107.2793.5963.9771. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.13b1ATPase Simulation - 327,506 AtomsEPYC 76012 x EPYC 76012 x Xeon Gold 6138246810Min: 0.91 / Avg: 0.91 / Max: 0.91Min: 0.45 / Avg: 0.45 / Max: 0.45Min: 0.66 / Avg: 0.66 / Max: 0.67

lzbench

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

OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: Brotli 0 - Process: CompressionEPYC 76012 x EPYC 76012 x Xeon Gold 613880160240320400SE +/- 0.88, N = 3SE +/- 0.88, N = 33443393581. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: Brotli 0 - Process: CompressionEPYC 76012 x EPYC 76012 x Xeon Gold 613860120180240300Min: 343 / Avg: 344.33 / Max: 346Min: 338 / Avg: 339.33 / Max: 3411. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: Brotli 0 - Process: DecompressionEPYC 76012 x EPYC 76012 x Xeon Gold 6138110220330440550SE +/- 2.00, N = 34034004861. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3
OpenBenchmarking.orgMB/s, More Is Betterlzbench 2017-08-08Test: Brotli 0 - Process: DecompressionEPYC 76012 x EPYC 76012 x Xeon Gold 613890180270360450Min: 399 / Avg: 403 / Max: 4051. (CXX) g++ options: -lrt -static -lpthread -fomit-frame-pointer -fstrict-aliasing -ffast-math -O3

FFTW

FFTW is a C subroutine library for computing the discrete Fourier transform (DFT) in one or more dimensions. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 613810002000300040005000SE +/- 26.81, N = 3SE +/- 25.22, N = 3SE +/- 22.91, N = 34408.004448.474534.701. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops Per Core, More Is BetterFFTW 3.3.6Performance Per Core - Build: Stock - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 6138306090120150137.7569.51113.371. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgMflops Per Thread, More Is BetterFFTW 3.3.6Performance Per Thread - Build: Stock - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 6138153045607568.8834.7556.681. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgMflops Per GHz, More Is BetterFFTW 3.3.6Performance Per Clock - Build: Stock - Size: 2D FFT Size 4096EPYC 76012 x Xeon Gold 61384008001200160020002003.641225.591. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgMflops Per Channel, More Is BetterFFTW 3.3.6Performance Per Memory Channel - Build: Stock - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 6138160320480640800551.00556.06755.781. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 61388001600240032004000Min: 4355.8 / Avg: 4408 / Max: 4444.7Min: 4407.6 / Avg: 4448.47 / Max: 4494.5Min: 4489.9 / Avg: 4534.7 / Max: 4565.41. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KSE +/- 124.60, N = 3SE +/- 41.43, N = 3SE +/- 200.20, N = 51287813332157021. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm
OpenBenchmarking.orgMflops Per Core, More Is BetterFFTW 3.3.6Performance Per Core - Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 613890180270360450402.44208.31392.551. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgMflops Per Thread, More Is BetterFFTW 3.3.6Performance Per Thread - Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 61384080120160200201.22104.16196.281. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgMflops Per GHz, More Is BetterFFTW 3.3.6Performance Per Clock - Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76012 x Xeon Gold 6138130026003900520065005853.644243.781. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgMflops Per Channel, More Is BetterFFTW 3.3.6Performance Per Memory Channel - Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 613860012001800240030001609.751666.502617.001. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Float + SSE - Size: 2D FFT Size 4096EPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KMin: 12633 / Avg: 12878 / Max: 13040Min: 13267 / Avg: 13332 / Max: 13409Min: 15370 / Avg: 15702.4 / Max: 164501. (CC) gcc options: -pthread -O3 -fomit-frame-pointer -mtune=native -malign-double -fstrict-aliasing -fno-schedule-insns -ffast-math -lm

Timed MAFFT Alignment

This test performs an alignment of 100 pyruvate decarboxylase sequences. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.392Multiple Sequence AlignmentEPYC 76012 x EPYC 76012 x Xeon Gold 61380.89781.79562.69343.59124.489SE +/- 0.08, N = 15SE +/- 0.03, N = 3SE +/- 0.04, N = 33.893.893.991. (CC) gcc options: -std=c99 -O3 -lm -lpthread
OpenBenchmarking.orgSeconds x Core, Fewer Is BetterTimed MAFFT Alignment 7.392Performance Per Core - Multiple Sequence AlignmentEPYC 76012 x EPYC 76012 x Xeon Gold 613850100150200250124.48248.96159.601. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgSeconds x Thread, Fewer Is BetterTimed MAFFT Alignment 7.392Performance Per Thread - Multiple Sequence AlignmentEPYC 76012 x EPYC 76012 x Xeon Gold 6138110220330440550248.96497.92319.201. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgSeconds x GHz, Fewer Is BetterTimed MAFFT Alignment 7.392Performance Per Clock - Multiple Sequence AlignmentEPYC 76012 x Xeon Gold 6138481216208.55814.7631. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgSeconds x Channel, Fewer Is BetterTimed MAFFT Alignment 7.392Performance Per Memory Channel - Multiple Sequence AlignmentEPYC 76012 x EPYC 76012 x Xeon Gold 613871421283531.1231.1223.941. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.392Multiple Sequence AlignmentEPYC 76012 x EPYC 76012 x Xeon Gold 6138246810Min: 3.44 / Avg: 3.89 / Max: 4.43Min: 3.83 / Avg: 3.89 / Max: 3.94Min: 3.92 / Avg: 3.99 / Max: 4.051. (CC) gcc options: -std=c99 -O3 -lm -lpthread

BLAKE2

This is a benchmark of BLAKE2 using the blake2s binary. BLAKE2 is a high-performance crypto alternative to MD5 and SHA-2/3. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgCycles Per Byte, Fewer Is BetterBLAKE2 20170307EPYC 76012 x EPYC 76012 x Xeon Gold 6138246810SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 136.866.863.131. (CC) gcc options: -O3 -march=native -lcrypto -lz
OpenBenchmarking.orgCycles Per Byte x Core, Fewer Is BetterBLAKE2 20170307Performance Per CoreEPYC 76012 x EPYC 76012 x Xeon Gold 6138100200300400500219.52439.04125.201. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgCycles Per Byte x Thread, Fewer Is BetterBLAKE2 20170307Performance Per ThreadEPYC 76012 x EPYC 76012 x Xeon Gold 61382004006008001000439.04878.08250.401. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgCycles Per Byte x GHz, Fewer Is BetterBLAKE2 20170307Performance Per ClockEPYC 76012 x Xeon Gold 61384812162015.0911.581. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgCycles Per Byte x Channel, Fewer Is BetterBLAKE2 20170307Performance Per Memory ChannelEPYC 76012 x EPYC 76012 x Xeon Gold 6138122436486054.8854.8818.781. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgCycles Per Byte, Fewer Is BetterBLAKE2 20170307EPYC 76012 x EPYC 76012 x Xeon Gold 61383691215Min: 6.86 / Avg: 6.86 / Max: 6.86Min: 6.85 / Avg: 6.86 / Max: 6.86Min: 3.1 / Avg: 3.13 / Max: 3.451. (CC) gcc options: -O3 -march=native -lcrypto -lz

GNU GMP GMPbench

GMPbench is a test of the GMP 6.1.2 math library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGMPbench Score, More Is BetterGNU GMP GMPbench 6.1.2Total TimeEPYC 76012 x EPYC 76012 x Xeon Gold 6138100020003000400050003949.903949.404731.501. (CC) gcc options: -O3 -fomit-frame-pointer -lm
OpenBenchmarking.orgGMPbench Score Per Core, More Is BetterGNU GMP GMPbench 6.1.2Performance Per Core - Total TimeEPYC 76012 x EPYC 76012 x Xeon Gold 6138306090120150123.4361.71118.291. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgGMPbench Score Per Thread, More Is BetterGNU GMP GMPbench 6.1.2Performance Per Thread - Total TimeEPYC 76012 x EPYC 76012 x Xeon Gold 6138142842567061.7230.8559.141. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgGMPbench Score Per GHz, More Is BetterGNU GMP GMPbench 6.1.2Performance Per Clock - Total TimeEPYC 76012 x Xeon Gold 61384008001200160020001795.411278.781. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgGMPbench Score Per Channel, More Is BetterGNU GMP GMPbench 6.1.2Performance Per Memory Channel - Total TimeEPYC 76012 x EPYC 76012 x Xeon Gold 61382004006008001000493.74493.68788.581. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6

Go Benchmarks

Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: jsonEPYC 76012 x EPYC 76012 x Xeon Gold 61381.4M2.8M4.2M5.6M7MSE +/- 35833.10, N = 3SE +/- 304763.07, N = 15SE +/- 12571.80, N = 3324827366314622637947
OpenBenchmarking.orgNanoseconds/Operation x Core, Fewer Is BetterGo BenchmarksPerformance Per Core - Test: jsonEPYC 76012 x EPYC 76012 x Xeon Gold 613890M180M270M360M450M1039447364244135681055178801. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgNanoseconds/Operation x Thread, Fewer Is BetterGo BenchmarksPerformance Per Thread - Test: jsonEPYC 76012 x EPYC 76012 x Xeon Gold 6138200M400M600M800M1000M2078894728488271362110357601. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgNanoseconds/Operation x GHz, Fewer Is BetterGo BenchmarksPerformance Per Clock - Test: jsonEPYC 76012 x Xeon Gold 61382M4M6M8M10M7146200.69760403.91. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgNanoseconds/Operation x Channel, Fewer Is BetterGo BenchmarksPerformance Per Memory Channel - Test: jsonEPYC 76012 x EPYC 76012 x Xeon Gold 613811M22M33M44M55M2598618453051696158276821. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: jsonEPYC 76012 x EPYC 76012 x Xeon Gold 61381.1M2.2M3.3M4.4M5.5MMin: 3189904 / Avg: 3248273 / Max: 3313469Min: 5403988 / Avg: 6631461.6 / Max: 9219823Min: 2614354 / Avg: 2637947 / Max: 2657272

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: buildEPYC 76012 x EPYC 76012 x Xeon Gold 61385000M10000M15000M20000M25000MSE +/- 45897387.75, N = 3SE +/- 68509188.98, N = 3SE +/- 64692103.72, N = 3228721150622202948362021421326863
OpenBenchmarking.orgNanoseconds/Operation x Core, Fewer Is BetterGo BenchmarksPerformance Per Core - Test: buildEPYC 76012 x EPYC 76012 x Xeon Gold 6138300000M600000M900000M1200000M1500000M73190768198414098869516808568530745201. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgNanoseconds/Operation x Thread, Fewer Is BetterGo BenchmarksPerformance Per Thread - Test: buildEPYC 76012 x EPYC 76012 x Xeon Gold 6138600000M1200000M1800000M2400000M3000000M1463815363968281977390336017137061490401. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgNanoseconds/Operation x GHz, Fewer Is BetterGo BenchmarksPerformance Per Clock - Test: buildEPYC 76012 x Xeon Gold 613820000M40000M60000M80000M100000M50318653136.479258909393.11. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgNanoseconds/Operation x Channel, Fewer Is BetterGo BenchmarksPerformance Per Memory Channel - Test: buildEPYC 76012 x EPYC 76012 x Xeon Gold 613840000M80000M120000M160000M200000M1829769204961762358689601285279611781. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: buildEPYC 76012 x EPYC 76012 x Xeon Gold 61384000M8000M12000M16000M20000MMin: 22790737605 / Avg: 22872115061.67 / Max: 22949586840Min: 21919397349 / Avg: 22029483620.33 / Max: 22155175794Min: 21321579237 / Avg: 21421326863 / Max: 21542566026

OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: garbageEPYC 76012 x EPYC 76012 x Xeon Gold 6138300K600K900K1200K1500KSE +/- 1487.41, N = 3SE +/- 18796.89, N = 5SE +/- 4798.77, N = 38749221536269857979
OpenBenchmarking.orgNanoseconds/Operation x Core, Fewer Is BetterGo BenchmarksPerformance Per Core - Test: garbageEPYC 76012 x EPYC 76012 x Xeon Gold 613820M40M60M80M100M2799750498321216343191601. EPYC 7601: Detected core count of 322. 2 x EPYC 7601: Detected core count of 643. 2 x Xeon Gold 6138: Detected core count of 40
OpenBenchmarking.orgNanoseconds/Operation x Thread, Fewer Is BetterGo BenchmarksPerformance Per Thread - Test: garbageEPYC 76012 x EPYC 76012 x Xeon Gold 613840M80M120M160M200M55995008196642432686383201. EPYC 7601: Detected thread count of 642. 2 x EPYC 7601: Detected thread count of 1283. 2 x Xeon Gold 6138: Detected thread count of 80
OpenBenchmarking.orgNanoseconds/Operation x GHz, Fewer Is BetterGo BenchmarksPerformance Per Clock - Test: garbageEPYC 76012 x Xeon Gold 6138700K1400K2100K2800K3500K1924828.43174522.31. EPYC 7601: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 2.202. 2 x Xeon Gold 6138: Detected GHz base clock speed (use PTS sensors for real-time frequency/sensor reporting) count of 3.70
OpenBenchmarking.orgNanoseconds/Operation x Channel, Fewer Is BetterGo BenchmarksPerformance Per Memory Channel - Test: garbageEPYC 76012 x EPYC 76012 x Xeon Gold 61383M6M9M12M15M69993761229015251478741. EPYC 7601: Detected channel count of 82. 2 x EPYC 7601: Detected channel count of 83. 2 x Xeon Gold 6138: Detected channel count of 6
OpenBenchmarking.orgNanoseconds/Operation, Fewer Is BetterGo BenchmarksTest: garbageEPYC 76012 x EPYC 76012 x Xeon Gold 6138300K600K900K1200K1500KMin: 873295 / Avg: 874921.67 / Max: 877892Min: 1477368 / Avg: 1536268.6 / Max: 1593887Min: 849199 / Avg: 857978.67 / Max: 865726

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2EPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KSE +/- 71.48, N = 20SE +/- 84.96, N = 20SE +/- 94.39, N = 7786157598690
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2EPYC 76012 x EPYC 76012 x Xeon Gold 613815003000450060007500Min: 7284 / Avg: 7860.55 / Max: 8354Min: 5031 / Avg: 5759.1 / Max: 6411Min: 8322 / Avg: 8690.43 / Max: 9058

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonEPYC 76012 x EPYC 76012 x Xeon Gold 613814002800420056007000SE +/- 51.84, N = 14SE +/- 73.31, N = 4SE +/- 17.77, N = 4652764664682
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: JythonEPYC 76012 x EPYC 76012 x Xeon Gold 613811002200330044005500Min: 6260 / Avg: 6526.71 / Max: 6994Min: 6268 / Avg: 6466 / Max: 6622Min: 4648 / Avg: 4681.5 / Max: 4717

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapEPYC 76012 x EPYC 76012 x Xeon Gold 613816003200480064008000SE +/- 155.25, N = 4SE +/- 267.63, N = 15SE +/- 87.69, N = 4701172957276
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradesoapEPYC 76012 x EPYC 76012 x Xeon Gold 613813002600390052006500Min: 6559 / Avg: 7010.75 / Max: 7268Min: 6411 / Avg: 7295.27 / Max: 10345Min: 7064 / Avg: 7275.75 / Max: 7450

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansEPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KSE +/- 158.29, N = 16SE +/- 170.94, N = 410418729312362
OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: TradebeansEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KMin: 6436 / Avg: 7293.31 / Max: 8855Min: 11900 / Avg: 12361.75 / Max: 12726

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Scala DottyEPYC 76012 x EPYC 76012 x Xeon Gold 61382K4K6K8K10KSE +/- 64.55, N = 40SE +/- 91.80, N = 40SE +/- 71.68, N = 109180.199580.117868.69
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Scala DottyEPYC 76012 x EPYC 76012 x Xeon Gold 613817003400510068008500Min: 8323.05 / Avg: 9180.19 / Max: 10026.35Min: 8788.64 / Avg: 9580.11 / Max: 11143.83Min: 7372.42 / Avg: 7868.69 / Max: 8179.92

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Apache Spark ALSEPYC 76012 x EPYC 76012 x Xeon Gold 613820K40K60K80K100KSE +/- 209.56, N = 32SE +/- 1657.87, N = 40SE +/- 170.86, N = 2520593.5979190.3628973.18
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Apache Spark ALSEPYC 76012 x EPYC 76012 x Xeon Gold 613814K28K42K56K70KMin: 18179.43 / Avg: 20593.59 / Max: 23902.63Min: 61982.41 / Avg: 79190.36 / Max: 101564Min: 26795.93 / Avg: 28973.18 / Max: 31109.21

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Apache Spark BayesEPYC 76012 x EPYC 76012 x Xeon Gold 613815003000450060007500SE +/- 72.10, N = 40SE +/- 393.53, N = 40SE +/- 31.05, N = 84995.836988.123880.90
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Apache Spark BayesEPYC 76012 x EPYC 76012 x Xeon Gold 613812002400360048006000Min: 3755.88 / Avg: 4995.83 / Max: 6332.1Min: 4572.85 / Avg: 6988.12 / Max: 15846.52Min: 3774.28 / Avg: 3880.9 / Max: 4039.69

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Savina Reactors.IOEPYC 76012 x EPYC 76012 x Xeon Gold 613810K20K30K40K50KSE +/- 413.08, N = 40SE +/- 801.91, N = 40SE +/- 523.32, N = 4027949.1544534.6829906.37
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Savina Reactors.IOEPYC 76012 x EPYC 76012 x Xeon Gold 61388K16K24K32K40KMin: 23219.77 / Avg: 27949.15 / Max: 34197.29Min: 36609.05 / Avg: 44534.68 / Max: 57132.98Min: 23963.37 / Avg: 29906.37 / Max: 36962.3

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Apache Spark PageRankEPYC 76012 x EPYC 76012 x Xeon Gold 61388K16K24K32K40KSE +/- 165.76, N = 40SE +/- 346.72, N = 33SE +/- 263.38, N = 828570.2435485.0625791.25
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Apache Spark PageRankEPYC 76012 x EPYC 76012 x Xeon Gold 61386K12K18K24K30KMin: 26368.43 / Avg: 28570.24 / Max: 31211.29Min: 32345.72 / Avg: 35485.06 / Max: 40846.9Min: 24613.71 / Avg: 25791.25 / Max: 27279.47

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: In-Memory Database ShootoutEPYC 76012 x EPYC 76012 x Xeon Gold 61384K8K12K16K20KSE +/- 103.07, N = 8SE +/- 185.20, N = 8SE +/- 179.83, N = 815230.5217582.8518103.14
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: In-Memory Database ShootoutEPYC 76012 x EPYC 76012 x Xeon Gold 61383K6K9K12K15KMin: 14855.37 / Avg: 15230.52 / Max: 15632.74Min: 16971.85 / Avg: 17582.85 / Max: 18525.27Min: 17484.61 / Avg: 18103.14 / Max: 19238.3

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Akka Unbalanced Cobwebbed TreeEPYC 76012 x EPYC 76012 x Xeon Gold 613811K22K33K44K55KSE +/- 847.65, N = 32SE +/- 517.04, N = 40SE +/- 176.93, N = 4028682.5652476.6525618.86
OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.9.0Test: Akka Unbalanced Cobwebbed TreeEPYC 76012 x EPYC 76012 x Xeon Gold 61389K18K27K36K45KMin: 25444.3 / Avg: 28682.56 / Max: 49888.3Min: 44634.61 / Avg: 52476.65 / Max: 58811.63Min: 23916.52 / Avg: 25618.86 / Max: 28557.79

Crafty

This is a performance test of Crafty, an advanced open-source chess engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeEPYC 76012 x EPYC 76012 x Xeon Gold 61381.6M3.2M4.8M6.4M8MSE +/- 859.26, N = 3SE +/- 12820.07, N = 3SE +/- 14207.53, N = 35730607568881173880991. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm
OpenBenchmarking.orgNodes Per Second Per Core, More Is BetterCrafty 25.2Performance Per Core - Elapsed TimeEPYC 76012 x EPYC 76012 x Xeon Gold 613840K80K120K160K200K179081.4788887.67184702.481. EPYC 7601: Detected core count of 322. 2 x