Clang znver1 Scheduler Model Tests x86-64 Haswell march

AMD Ryzen 7 1700 Eight-Core testing with the new Zen znver1 scheduler model. Tests by Michael Larabel for a future article.

HTML result view exported from: https://openbenchmarking.org/result/1707201-TR-CLANGZNVE41&grr&rdt.

Clang znver1 Scheduler Model Tests x86-64 Haswell marchProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay DriverCompilerFile-SystemScreen Resolution-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswellAMD Ryzen 7 1700 Eight-Core @ 3.00GHz (16 Cores)MSI B350 TOMAHAWK (MS-7A34) v1.0AMD Device 145016384MB120GB Samsung SSD 840HIS AMD Radeon HD 7750/8740 / R7 250E 1024MBAMD Cape Verde/PitcairnRealtek RTL8111/8168/8411Ubuntu 17.044.12.0-999-generic (x86_64) 20170622Unity 7.5.0modesetting 1.19.3Clang 6.0.0 (SVN 308494)ext41024x7684.13.0-999-generic (x86_64) 20170622OpenBenchmarking.orgCompiler Details- Optimized build; Default target: x86_64-unknown-linux-gnu; Host CPU: znver1Processor Details- Scaling Governor: acpi-cpufreq ondemand

Clang znver1 Scheduler Model Tests x86-64 Haswell marchapache: Static Web Page Servinghint: FLOATredis: SETredis: GETredis: LPOPpgbench: Buffer Test - Heavy Contention - Read Writepgbench: Buffer Test - Single Thread - Read Writepgbench: Buffer Test - Normal Load - Read Writetjbench: Decompression Throughputtachyon: Total Timeencode-wavpack: WAV To WavPackencode-mp3: WAV To MP3encode-flac: WAV To FLACbullet: Convex Trimeshbullet: Prim Trimeshbullet: 136 Ragdollsbullet: 1000 Convexbullet: 1000 Stackbullet: 3000 Fallbullet: Raytestsstockfish: Total Timec-ray: Total Timeebizzy: himeno: Poisson Pressure Solvergraphics-magick: Local Adaptive Thresholdinggraphics-magick: HWB Color Spacegraphics-magick: Sharpengraphics-magick: Blurx264: H.264 Video Encodingtscp: AI Chess Performancescimark2: Jacobi Successive Over-Relaxationscimark2: Dense LU Matrix Factorizationscimark2: Sparse Matrix Multiplyscimark2: Fast Fourier Transformscimark2: Monte Carloscimark2: Compositefftw: Float + SSE - 2D FFT Size 4096-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell26994.80301795000.851385134.041953368.381968693.381938.93226.041865.09161.785.296.529.756.771.201.002.964.845.274.622.86371013.684085091110.92133173137136252.3311033311675.035572.722623.75133.36642.552129.481369027700.93300351844.051413713.291973143.291960971.501875.27224.891813.34165.305.296.4510.925.691.160.992.934.645.104.532.78367013.623829601022.65136178139140252.5210963031679.215692.832597.54134.24657.962152.361339826277.48299157100.641411858.831916990.882004499.671488.59216.831662.61167.205.306.5010.955.671.120.982.934.475.204.572.72374513.774007651123.43126176139132233.9210902231673.655454.002627.46240.21653.542129.7713133OpenBenchmarking.org

Apache Benchmark

Static Web Page Serving

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page Serving-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell6K12K18K24K30KSE +/- 179.13, N = 3SE +/- 44.40, N = 3SE +/- 337.31, N = 326994.8027700.9326277.48-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -shared -fPIC -pthread -O3

Hierarchical INTegration

Test: FLOAT

OpenBenchmarking.orgQUIPs, More Is BetterHierarchical INTegration 1.0Test: FLOAT-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell60M120M180M240M300MSE +/- 120507.74, N = 3SE +/- 20484.23, N = 3SE +/- 98377.24, N = 3301795000.85300351844.05299157100.64-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

Redis

Test: SET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: SET-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell300K600K900K1200K1500KSE +/- 8018.20, N = 3SE +/- 23768.12, N = 6SE +/- 8142.48, N = 31385134.041413713.291411858.831. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: GET-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell400K800K1200K1600K2000KSE +/- 15420.28, N = 3SE +/- 27406.80, N = 3SE +/- 7458.07, N = 31953368.381973143.291916990.881. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl

Redis

Test: LPOP

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: LPOP-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell400K800K1200K1600K2000KSE +/- 13701.81, N = 3SE +/- 13594.16, N = 3SE +/- 36043.81, N = 61968693.381960971.502004499.671. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl

PostgreSQL pgbench

Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 9.6.3Scaling: Buffer Test - Test: Heavy Contention - Mode: Read Write-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell400800120016002000SE +/- 30.24, N = 3SE +/- 30.52, N = 4SE +/- 97.18, N = 61938.931875.271488.59-march=x86-64-march=znver1-march=haswell -lpgport1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -fpic -shared

PostgreSQL pgbench

Scaling: Buffer Test - Test: Single Thread - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 9.6.3Scaling: Buffer Test - Test: Single Thread - Mode: Read Write-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell50100150200250SE +/- 0.39, N = 3SE +/- 1.25, N = 3SE +/- 2.56, N = 3226.04224.89216.83-march=x86-64-march=znver1-march=haswell -lpgport1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -fpic -shared

PostgreSQL pgbench

Scaling: Buffer Test - Test: Normal Load - Mode: Read Write

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 9.6.3Scaling: Buffer Test - Test: Normal Load - Mode: Read Write-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell400800120016002000SE +/- 47.88, N = 6SE +/- 52.04, N = 6SE +/- 96.50, N = 61865.091813.341662.61-march=x86-64-march=znver1-march=haswell -lpgport1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -fpic -shared

libjpeg-turbo tjbench

Test: Decompression Throughput

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 1.5.1Test: Decompression Throughput-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell4080120160200SE +/- 0.11, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 3161.78165.30167.20-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

Tachyon

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterTachyon 0.98.9Total Time-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell1.19252.3853.57754.775.9625SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 35.295.295.301. (CC) gcc options: -m32 -O3 -fomit-frame-pointer -ffast-math -ltachyon -lm -lpthread

WavPack Audio Encoding

WAV To WavPack

OpenBenchmarking.orgSeconds, Fewer Is BetterWavPack Audio Encoding 5.1WAV To WavPack-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell246810SE +/- 0.01, N = 5SE +/- 0.00, N = 5SE +/- 0.06, N = 56.526.456.50-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

LAME MP3 Encoding

WAV To MP3

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.99.3WAV To MP3-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell3691215SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 59.7510.9210.95-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -ffast-math -funroll-loops -fschedule-insns2 -fbranch-count-reg -fforce-addr -pipe -lm

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.1WAV To FLAC-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell246810SE +/- 0.00, N = 5SE +/- 0.00, N = 5SE +/- 0.00, N = 56.775.695.67-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -logg -lm

Bullet Physics Engine

Test: Convex Trimesh

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Convex Trimesh-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell0.270.540.811.081.35SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.201.161.12-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: Prim Trimesh

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Prim Trimesh-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell0.2250.450.6750.91.125SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.000.990.98-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 136 Ragdolls

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 136 Ragdolls-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell0.6661.3321.9982.6643.33SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.962.932.93-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 1000 Convex

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 1000 Convex-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell1.0892.1783.2674.3565.445SE +/- 0.00, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 34.844.644.47-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 1000 Stack

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 1000 Stack-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell1.18582.37163.55744.74325.929SE +/- 0.01, N = 3SE +/- 0.01, N = 3SE +/- 0.01, N = 35.275.105.20-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 3000 Fall

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 3000 Fall-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell1.03952.0793.11854.1585.1975SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 34.624.534.57-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: Raytests

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Raytests-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell0.64351.2871.93052.5743.2175SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 32.862.782.72-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -O3 -rdynamic -lglut -lGL -lGLU

Stockfish

Total Time

OpenBenchmarking.orgms, Fewer Is BetterStockfish 2014-11-26Total Time-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell8001600240032004000SE +/- 1.76, N = 3SE +/- 5.03, N = 3SE +/- 3.51, N = 3371036703745-march=x86-64-march=znver1-march=haswell1. (CXX) g++ options: -lpthread -O3 -fno-exceptions -fno-rtti -ansi -pedantic -msse -msse3 -mpopcnt

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell48121620SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.01, N = 313.6813.6213.77-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -lm -lpthread -O3

ebizzy

OpenBenchmarking.orgRecords/s, More Is Betterebizzy 0.3-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell90K180K270K360K450KSE +/- 4432.38, N = 3SE +/- 5536.66, N = 6SE +/- 7907.66, N = 6408509382960400765-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -pthread -lpthread -O3

Himeno Benchmark

Poisson Pressure Solver

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure Solver-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell2004006008001000SE +/- 1.19, N = 3SE +/- 5.42, N = 3SE +/- 0.41, N = 31110.921022.651123.43-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -mavx2

GraphicsMagick

Operation: Local Adaptive Thresholding

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Local Adaptive Thresholding-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell306090120150133136126-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lgomp -lpthread

GraphicsMagick

Operation: HWB Color Space

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: HWB Color Space-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell4080120160200SE +/- 0.33, N = 3173178176-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lgomp -lpthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Sharpen-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell306090120150SE +/- 0.33, N = 3137139139-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lgomp -lpthread

GraphicsMagick

Operation: Blur

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: Blur-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell306090120150136140132-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -fopenmp -O3 -pthread -ljbig -lwebp -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lgomp -lpthread

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2017-05-19H.264 Video Encoding-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell60120180240300SE +/- 0.34, N = 5SE +/- 0.68, N = 5SE +/- 1.29, N = 5252.33252.52233.92-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fomit-frame-pointer -fno-tree-vectorize

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess Performance-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell200K400K600K800K1000KSE +/- 1625.76, N = 5SE +/- 1629.46, N = 5SE +/- 1257.19, N = 5110333110963031090223-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3

SciMark

Computational Test: Jacobi Successive Over-Relaxation

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-Relaxation-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell400800120016002000SE +/- 0.35, N = 4SE +/- 0.50, N = 4SE +/- 1.64, N = 41675.031679.211673.65-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Dense LU Matrix Factorization

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix Factorization-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell12002400360048006000SE +/- 14.40, N = 4SE +/- 16.27, N = 4SE +/- 22.62, N = 45572.725692.835454.00-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Sparse Matrix Multiply

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix Multiply-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell6001200180024003000SE +/- 4.33, N = 4SE +/- 5.05, N = 4SE +/- 8.87, N = 42623.752597.542627.46-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Fast Fourier Transform

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier Transform-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell50100150200250SE +/- 0.55, N = 4SE +/- 0.63, N = 4SE +/- 0.28, N = 4133.36134.24240.21-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Monte Carlo

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte Carlo-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell140280420560700SE +/- 0.08, N = 4SE +/- 0.67, N = 4SE +/- 0.24, N = 4642.55657.96653.54-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

SciMark

Computational Test: Composite

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Composite-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell5001000150020002500SE +/- 2.34, N = 4SE +/- 2.54, N = 4SE +/- 5.26, N = 42129.482152.362129.77-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

FFTW

Build: Float + SSE - Size: 2D FFT Size 4096

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.4Build: Float + SSE - Size: 2D FFT Size 4096-O3 -march=x86-64-O3 -march=znver1-O3 -march=haswell3K6K9K12K15KSE +/- 12.52, N = 5SE +/- 55.71, N = 5SE +/- 51.03, N = 5136901339813133-march=x86-64-march=znver1-march=haswell1. (CC) gcc options: -O3 -lm

Phoronix Test Suite v10.8.5