AMD EPYC Compiler Testing July 2018

AMD EPYC compiler benchmarks.

HTML result view exported from: https://openbenchmarking.org/result/1807201-RA-AMDEPYCCO18.

AMD EPYC Compiler Testing July 2018ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelCompilerFile-SystemScreen ResolutionGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-CoreAMD EPYC 7601 32-Core @ 2.20GHz (32 Cores / 64 Threads)TYAN B8026T70AE24HR (V1.02.B10 BIOS)AMD Family 17h129024MB280GB INTEL SSDPE21D280GAASPEED ASPEED FamilyVE228Broadcom Limited NetXtreme BCM5720 Gigabit PCIeUbuntu 18.104.16.0-041600-generic (x86_64)GCC 8.1.1 20180720ext41920x1080OpenBenchmarking.orgEnvironment Details- CXXFLAGS=-O3-march=native CFLAGS=-O3-march=nativeCompiler Details- --disable-multilib --enable-checking=releaseDisk Details- NONE / data=ordered,errors=remount-ro,relatime,rwProcessor Details- Scaling Governor: acpi-cpufreq ondemandSecurity Details- __user pointer sanitization + Full AMD retpoline IBPB Protection

AMD EPYC Compiler Testing July 2018scimark2: Compositescimark2: Fast Fourier Transformscimark2: Jacobi Successive Over-Relaxationscimark2: Monte Carloscimark2: Sparse Matrix Multiplyscimark2: Dense LU Matrix Factorizationvpxenc: vpxencfftw: Stock - 1D FFT Size 1024fftw: Stock - 2D FFT Size 1024polybench-c: 3 Matrix Multiplicationspolybench-c: Correlation Computationpolybench-c: Covariance Computationsqlite: Timed SQLite Insertionsopenssl: RSA 4096-bit Performanceaobench: 2048 x 2048 - Total Timecompress-7zip: Compress Speed Testcompress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19m-queens: Time To Solvec-ray: Total Timecrafty: Elapsed Timesmallpt: Global Illumination Renderer; 100 Samplesstockfish: Total Timetscp: AI Chess Performanceredis: SETredis: GETredis: LPUSHredis: LPOPredis: SADDpgbench: Buffer Test - Single Thread - Read Writepgbench: Buffer Test - Single Thread - Read Onlypgbench: Buffer Test - Normal Load - Read Writepgbench: Buffer Test - Normal Load - Read Onlytjbench: Decompression Throughputapache: Static Web Page Servingencode-mp3: WAV To MP3encode-flac: WAV To FLACebizzy: graphics-magick: HWB Color Spacegraphics-magick: Blurgraphics-magick: Local Adaptive Thresholdinggraphics-magick: Resizinggraphics-magick: Sharpenhmmer: Pfam Database Searchbullet: 3000 Fallbullet: 1000 Stackbullet: 136 Ragdollsbullet: 1000 Convexbullet: Prim Trimeshbullet: Convex Trimeshbullet: Rayteststtsiod-renderer: Phong Rendering With Soft-Shadow Mappinghimeno: Poisson Pressure Solvertinymembench: Standard Memcpysockperf: Latency Under Loadsockperf: Latency Ping PongGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core1943.30219.381687.64556.372421.384831.7512.388518.676557.833.585.345.342.894610.6350.80100684155.6432.022.5260148664597302908510191375161.961691908.001108304.541108268.081430638.452071.7814374.1336226.69408873.25142.1816205.3210.5112.6311470092091541101851887.655.025.793.095.321.131.333.11439.54982.6712497.2019.756.421766.38214.571676.49195.982442.874301.998289.106351.503.665.395.373.134656.4057.8598726260.022694.47566.22599875953313725107295071427.389213.4528411.49242221.35106.0315210.4715.4012.871131696461823296192.263.04972.0712463.5031.296.88OpenBenchmarking.org

SciMark

Computational Test: Composite

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: CompositeGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core400800120016002000SE +/- 5.96, N = 31943.301766.381. (CC) gcc options: -O3 -march=native -lm

SciMark

Computational Test: Fast Fourier Transform

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core50100150200250SE +/- 0.10, N = 3219.38214.571. (CC) gcc options: -O3 -march=native -lm

SciMark

Computational Test: Jacobi Successive Over-Relaxation

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core400800120016002000SE +/- 0.06, N = 31687.641676.491. (CC) gcc options: -O3 -march=native -lm

SciMark

Computational Test: Monte Carlo

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core120240360480600SE +/- 0.10, N = 3556.37195.981. (CC) gcc options: -O3 -march=native -lm

SciMark

Computational Test: Sparse Matrix Multiply

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core5001000150020002500SE +/- 11.03, N = 32421.382442.871. (CC) gcc options: -O3 -march=native -lm

SciMark

Computational Test: Dense LU Matrix Factorization

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core10002000300040005000SE +/- 20.82, N = 34831.754301.991. (CC) gcc options: -O3 -march=native -lm

VP9 libvpx Encoding

vpxenc

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.7.0vpxencGCC 8.2 RC1: -O3 -march=native3691215SE +/- 0.02, N = 312.381. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE

FFTW

Build: Stock - Size: 1D FFT Size 1024

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 1D FFT Size 1024GCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core2K4K6K8K10KSE +/- 18.71, N = 38518.678289.101. (CC) gcc options: -pthread -O3 -march=native -lm

FFTW

Build: Stock - Size: 2D FFT Size 1024

OpenBenchmarking.orgMflops, More Is BetterFFTW 3.3.6Build: Stock - Size: 2D FFT Size 1024GCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core14002800420056007000SE +/- 73.40, N = 36557.836351.501. (CC) gcc options: -pthread -O3 -march=native -lm

PolyBench-C

Test: 3 Matrix Multiplications

OpenBenchmarking.orgSeconds, Fewer Is BetterPolyBench-C 4.2Test: 3 Matrix MultiplicationsGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core0.82351.6472.47053.2944.1175SE +/- 0.01, N = 33.583.661. (CC) gcc options: -O3 -march=native

PolyBench-C

Test: Correlation Computation

OpenBenchmarking.orgSeconds, Fewer Is BetterPolyBench-C 4.2Test: Correlation ComputationGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core1.21282.42563.63844.85126.064SE +/- 0.01, N = 35.345.391. (CC) gcc options: -O3 -march=native

PolyBench-C

Test: Covariance Computation

OpenBenchmarking.orgSeconds, Fewer Is BetterPolyBench-C 4.2Test: Covariance ComputationGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core1.20832.41663.62494.83326.0415SE +/- 0.02, N = 35.345.371. (CC) gcc options: -O3 -march=native

SQLite

Timed SQLite Insertions

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.22Timed SQLite InsertionsGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core0.70431.40862.11292.81723.5215SE +/- 0.03, N = 32.893.131. (CC) gcc options: -O3 -march=native -lz -ldl -lpthread

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.0fRSA 4096-bit PerformanceGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core10002000300040005000SE +/- 21.56, N = 34610.634656.40-lssl1. (CC) gcc options: -O3 -pthread -m64 -lcrypto -ldl

AOBench

Size: 2048 x 2048 - Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterAOBenchSize: 2048 x 2048 - Total TimeGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core1326395265SE +/- 0.01, N = 350.8057.851. (CC) gcc options: -lm -O3 -march=native

7-Zip Compression

Compress Speed Test

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 16.02Compress Speed TestGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core20K40K60K80K100KSE +/- 715.28, N = 3100684987261. (CXX) g++ options: -pipe -lpthread

Zstd Compression

Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19

OpenBenchmarking.orgSeconds, Fewer Is BetterZstd Compression 1.3.4Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19GCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core60120180240300SE +/- 0.54, N = 3155.64260.021. (CC) gcc options: -O3 -march=native -pthread -lz -llzma

m-queens

Time To Solve

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.1Time To SolveGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core6001200180024003000SE +/- 0.14, N = 332.022694.471. (CXX) g++ options: -fopenmp -O3 -march=native -O2

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core120240360480600SE +/- 0.01, N = 32.52566.221. (CC) gcc options: -lm -lpthread -O3 -march=native

Crafty

Elapsed Time

OpenBenchmarking.orgNodes Per Second, More Is BetterCrafty 25.2Elapsed TimeGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core1.3M2.6M3.9M5.2M6.5MSE +/- 5146.22, N = 3601486659987591. (CC) gcc options: -pthread -lstdc++ -fprofile-use -lm

Smallpt

Global Illumination Renderer; 100 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core12024036048060045331. (CXX) g++ options: -fopenmp -O3 -march=native

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 9Total TimeGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core13M26M39M52M65MSE +/- 576746.80, N = 35973029013725101. (CXX) g++ options: -m64 -lpthread -O3 -march=native -fno-exceptions -std=c++11 -pedantic -msse -msse3 -mpopcnt -flto

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core200K400K600K800K1000KSE +/- 766.17, N = 58510197295071. (CC) gcc options: -O3 -march=native

Redis

Test: SET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: SETGCC 8.2 RC1: -O3 -march=native300K600K900K1200K1500KSE +/- 13759.71, N = 31375161.961. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: GET

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: GETGCC 8.2 RC1: -O3 -march=native400K800K1200K1600K2000KSE +/- 37527.46, N = 61691908.001. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: LPUSH

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: LPUSHGCC 8.2 RC1: -O3 -march=native200K400K600K800K1000KSE +/- 18347.41, N = 61108304.541. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: LPOP

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: LPOPGCC 8.2 RC1: -O3 -march=native200K400K600K800K1000KSE +/- 4079.04, N = 31108268.081. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

Redis

Test: SADD

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 4.0.8Test: SADDGCC 8.2 RC1: -O3 -march=native300K600K900K1200K1500KSE +/- 4081.67, N = 31430638.451. (CC) gcc options: -ggdb -rdynamic -lm -ldl -pthread

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Single Thread - Mode: Read WriteGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core400800120016002000SE +/- 12.52, N = 32071.781427.38-lpq1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Single Thread - Mode: Read OnlyGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core3K6K9K12K15KSE +/- 113.33, N = 314374.139213.45-lpq1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core8K16K24K32K40KSE +/- 133.09, N = 336226.6928411.49-lpq1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm

PostgreSQL pgbench

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

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 10.3Scaling: Buffer Test - Test: Normal Load - Mode: Read OnlyGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core90K180K270K360K450KSE +/- 1289.79, N = 3408873.25242221.35-lpq1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpthread -lrt -lcrypt -ldl -lm

libjpeg-turbo tjbench

Test: Decompression Throughput

OpenBenchmarking.orgMegapixels/sec, More Is Betterlibjpeg-turbo tjbench 1.5.3Test: Decompression ThroughputGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core306090120150SE +/- 0.01, N = 3142.18106.031. (CC) gcc options: -O3 -march=native -lm

Apache Benchmark

Static Web Page Serving

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page ServingGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core3K6K9K12K15KSE +/- 193.23, N = 316205.3215210.471. (CC) gcc options: -shared -fPIC -pthread -O3 -march=native

LAME MP3 Encoding

WAV To MP3

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.100WAV To MP3GCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core48121620SE +/- 0.00, N = 310.5115.401. (CC) gcc options: -O3 -march=native -lm

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.2WAV To FLACGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core3691215SE +/- 0.19, N = 512.6312.871. (CXX) g++ options: -O3 -march=native -fvisibility=hidden -logg -lm

ebizzy

OpenBenchmarking.orgRecords/s, More Is Betterebizzy 0.3GCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core200K400K600K800K1000KSE +/- 28582.82, N = 6114700911316961. (CC) gcc options: -pthread -lpthread -O3 -march=native

GraphicsMagick

Operation: HWB Color Space

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.28Operation: HWB Color SpaceGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core50100150200250209461. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread

GraphicsMagick

Operation: Blur

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.28Operation: BlurGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core306090120150154181. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread

GraphicsMagick

Operation: Local Adaptive Thresholding

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.28Operation: Local Adaptive ThresholdingGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core20406080100110231. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread

GraphicsMagick

Operation: Resizing

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.28Operation: ResizingGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core4080120160200SE +/- 1.20, N = 3185291. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread

GraphicsMagick

Operation: Sharpen

OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.28Operation: SharpenGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core408012016020018861. (CC) gcc options: -fopenmp -O3 -march=native -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -ldl -lpthread

Timed HMMer Search

Pfam Database Search

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 2.3.2Pfam Database SearchGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core4080120160200SE +/- 0.05, N = 37.65192.261. (CC) gcc options: -O3 -march=native -pthread -lhmmer -lsquid -lm

Bullet Physics Engine

Test: 3000 Fall

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 3000 FallGCC 8.2 RC1: -O3 -march=native1.12952.2593.38854.5185.6475SE +/- 0.00, N = 35.021. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 1000 Stack

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 1000 StackGCC 8.2 RC1: -O3 -march=native1.30282.60563.90845.21126.514SE +/- 0.01, N = 35.791. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 136 Ragdolls

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 136 RagdollsGCC 8.2 RC1: -O3 -march=native0.69531.39062.08592.78123.4765SE +/- 0.00, N = 33.091. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: 1000 Convex

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: 1000 ConvexGCC 8.2 RC1: -O3 -march=native1.1972.3943.5914.7885.985SE +/- 0.01, N = 35.321. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: Prim Trimesh

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Prim TrimeshGCC 8.2 RC1: -O3 -march=native0.25430.50860.76291.01721.2715SE +/- 0.00, N = 31.131. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: Convex Trimesh

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: Convex TrimeshGCC 8.2 RC1: -O3 -march=native0.29930.59860.89791.19721.4965SE +/- 0.00, N = 31.331. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

Bullet Physics Engine

Test: Raytests

OpenBenchmarking.orgSeconds, Fewer Is BetterBullet Physics Engine 2.81Test: RaytestsGCC 8.2 RC1: -O3 -march=native0.69981.39962.09942.79923.499SE +/- 0.00, N = 33.111. (CXX) g++ options: -O3 -march=native -rdynamic -lglut -lGL -lGLU

TTSIOD 3D Renderer

Phong Rendering With Soft-Shadow Mapping

OpenBenchmarking.orgFPS, More Is BetterTTSIOD 3D Renderer 2.3aPhong Rendering With Soft-Shadow MappingGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core100200300400500SE +/- 1.94, N = 3439.543.041. (CXX) g++ options: -O3 -march=native -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -lstdc++

Himeno Benchmark

Poisson Pressure Solver

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core2004006008001000SE +/- 0.78, N = 3982.67972.071. (CC) gcc options: -O3 -march=native -mavx2

Tinymembench

Standard Memcpy

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemcpyGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core3K6K9K12K15K12497.2012463.501. (CC) gcc options: -O2 -O3 -march=native -lm

Tinymembench

Standard Memset

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemsetGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core2K4K6K8K10K10063.1010015.501. (CC) gcc options: -O2 -O3 -march=native -lm

Sockperf

Test: Latency Under Load

OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.4Test: Latency Under LoadGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core714212835SE +/- 0.30, N = 1019.7531.291. (CXX) g++ options: --param -O3 -march=native -rdynamic -ldl -lpthread

Sockperf

Test: Latency Ping Pong

OpenBenchmarking.orgusec, Fewer Is BetterSockperf 3.4Test: Latency Ping PongGCC 8.2 RC1: -O3 -march=nativeAMD EPYC 7601 32-Core246810SE +/- 0.11, N = 106.426.881. (CXX) g++ options: --param -O3 -march=native -rdynamic -ldl -lpthread

Phoronix Test Suite v10.8.4