Threadripper 2990WX Thread Scaling

AMD Ryzen Threadripper 2990WX 32-Core testing with different core/thread counts. Benchmarks by Michael Larabel for a future article on Phoronix.

HTML result view exported from: https://openbenchmarking.org/result/1808163-RA-TR2SCALIN95&grw&rdt.

Threadripper 2990WX Thread ScalingProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen Resolution64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 ThreadAMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores / 64 Threads)ASUS ROG ZENITH EXTREME (0064 BIOS)AMD Family 17h32768MBSamsung SSD 970 EVO 500GBAMD Radeon RX Vega 8192MBRealtek ALC1220DELL P2415QIntel I211 Gigabit Connection + Qualcomm Atheros QCA6174 802.11ac WirelessUbuntu 18.044.18.0-999-generic (x86_64) 20180805GNOME Shell 3.28.2X Server 1.19.6amdgpu 18.0.14.5 Mesa 18.0.5 (LLVM 6.0.0)GCC 7.3.0ext43840x2160AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (32 Cores)AMD Radeon RX Vega 8064MBAMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (16 Cores)AMD Radeon RX Vega 8192MBAMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (8 Cores)AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (4 Cores)AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (2 Cores)AMD Ryzen Threadripper 2990WX 32-Core @ 3.00GHz (1 Core)OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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-as=/usr/bin/x86_64-linux-gnu-as --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-ld=/usr/bin/x86_64-linux-gnu-ld --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic --without-cuda-driver -v Processor Details- Scaling Governor: acpi-cpufreq ondemandGraphics Details- GLAMORSecurity Details- __user pointer sanitization + Full AMD retpoline IBPB + SSB disabled via prctl and seccomp Protection

Threadripper 2990WX Thread Scalingnpb: EP.Cstockfish: Total Timejohn-the-ripper: Blowfishasmfish: 1024 Hash Memory, 26 Depthm-queens: Time To Solvebuild-linux-kernel: Time To Compilesysbench: CPUc-ray: Total Time - 4K, 16 Rays Per Pixelv-ray: CPUblender: Barbershop - CPU-Onlypovray: Trace Timeindigobench: Bedroomindigobench: Supercaropenssl: RSA 4096-bit Performanceebizzy: 64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread2160.7566760558446827672805724.7931.7256727.7824.6330.75402.2517.383.076.565869.6313677171092.6553378671309115905866336.3435.2553346.1925.5635.90553.2422.512.575.675514.601555472594.5029758991148103295562365.3054.3029095.4346.5960.26982.1639.181.433.163002.271145295313.3216545321881118055501121.7892.0315350.9388.31108.601801.1873.360.781.741582.67671058160.75896390660369672981235.56171.227880.26172.28204.043276.02141.320.430.97811.3334346284.13464735133065153369447.21320.884136.77327.35381.55267.950.230.51422.7317769442.08234176416602589854892.50625.132065.06655.74761.83538.040.110.25212.6790996OpenBenchmarking.org

NAS Parallel Benchmarks

Test / Class: EP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: EP.C64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread5001000150020002500SE +/- 1.71, N = 3SE +/- 0.17, N = 3SE +/- 0.04, N = 3SE +/- 0.09, N = 3SE +/- 0.02, N = 3SE +/- 0.29, N = 3SE +/- 0.12, N = 32160.751092.65594.50313.32160.7584.1342.081. (F9X) gfortran options: -O3 -march=native -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi2. Open MPI 2.1.1

Stockfish

Total Time

OpenBenchmarking.orgNodes Per Second, More Is BetterStockfish 9Total Time64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread14M28M42M56M70MSE +/- 651663.27, N = 3SE +/- 217995.10, N = 3SE +/- 86990.41, N = 3SE +/- 86158.77, N = 3SE +/- 71881.58, N = 3SE +/- 21923.77, N = 3SE +/- 2751.54, N = 3667605585337867129758991165453218963906464735123417641. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -msse -msse3 -mpopcnt -flto

John The Ripper

Test: Blowfish

OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0-jumbo-1Test: Blowfish64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread10K20K30K40K50KSE +/- 245.29, N = 3SE +/- 565.26, N = 3SE +/- 14.17, N = 3SE +/- 372.22, N = 6SE +/- 74.63, N = 3SE +/- 14.73, N = 3SE +/- 18.33, N = 344682309111481088116036330616601. (CC) gcc options: -lssl -lcrypto -fopenmp -pthread -lm -lz -ldl -lcrypt -lbz2

asmFish

1024 Hash Memory, 26 Depth

OpenBenchmarking.orgNodes/second, More Is BetterasmFish 2017-09-191024 Hash Memory, 26 Depth64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread16M32M48M64M80MSE +/- 298662.02, N = 3SE +/- 116017.90, N = 3SE +/- 313307.04, N = 3SE +/- 133076.25, N = 3SE +/- 67053.30, N = 3SE +/- 44526.21, N = 3SE +/- 619.33, N = 376728057590586633295562318055501967298151533692589854

m-queens

Time To Solve

OpenBenchmarking.orgSeconds, Fewer Is Betterm-queens 1.1Time To Solve64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread2004006008001000SE +/- 0.10, N = 3SE +/- 0.00, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 3SE +/- 0.22, N = 3SE +/- 0.27, N = 324.7936.3465.30121.78235.56447.21892.501. (CXX) g++ options: -fopenmp -O2 -march=native

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.13Time To Compile64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread130260390520650SE +/- 0.43, N = 6SE +/- 0.47, N = 6SE +/- 0.87, N = 3SE +/- 0.98, N = 3SE +/- 0.92, N = 3SE +/- 0.79, N = 3SE +/- 0.88, N = 331.7235.2554.3092.03171.22320.88625.13

Sysbench

Test: CPU

OpenBenchmarking.orgEvents Per Second, More Is BetterSysbench 2018-07-28Test: CPU64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread12K24K36K48K60KSE +/- 65.28, N = 3SE +/- 32.79, N = 3SE +/- 8.26, N = 3SE +/- 5.20, N = 3SE +/- 1.58, N = 3SE +/- 0.73, N = 3SE +/- 0.87, N = 356727.7853346.1929095.4315350.937880.264136.772065.061. (CC) gcc options: -pthread -O3 -funroll-loops -ggdb3 -march=amdfam10 -rdynamic -ldl -laio -lm

C-Ray

Total Time - 4K, 16 Rays Per Pixel

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total Time - 4K, 16 Rays Per Pixel64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread140280420560700SE +/- 0.06, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.29, N = 3SE +/- 0.13, N = 3SE +/- 0.65, N = 324.6325.5646.5988.31172.28327.35655.741. (CC) gcc options: -lm -lpthread -O3

Chaos Group V-RAY

Mode: CPU

OpenBenchmarking.orgSeconds, Fewer Is BetterChaos Group V-RAY 1.1.0Mode: CPU64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread160320480640800SE +/- 0.11, N = 3SE +/- 0.09, N = 3SE +/- 0.16, N = 3SE +/- 0.34, N = 3SE +/- 0.04, N = 3SE +/- 1.08, N = 3SE +/- 1.89, N = 330.7535.9060.26108.60204.04381.55761.83

Blender

Blend File: Barbershop - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 2.79aBlend File: Barbershop - Compute: CPU-Only64 Threads32 Threads16 Threads8 Threads4 Threads7001400210028003500402.25553.24982.161801.183276.02

POV-Ray

Trace Time

OpenBenchmarking.orgSeconds, Fewer Is BetterPOV-Ray 3.7.0.7Trace Time64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread120240360480600SE +/- 0.05, N = 3SE +/- 0.15, N = 3SE +/- 0.02, N = 3SE +/- 0.43, N = 3SE +/- 0.63, N = 3SE +/- 0.46, N = 3SE +/- 0.57, N = 317.3822.5139.1873.36141.32267.95538.041. (CXX) g++ options: -pipe -O3 -ffast-math -march=native -pthread -lSDL -lSM -lICE -lX11 -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system

IndigoBench

Scene: Bedroom

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.0.64Scene: Bedroom64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread0.69081.38162.07242.76323.454SE +/- 0.12, N = 6SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 33.072.571.430.780.430.230.11

IndigoBench

Scene: Supercar

OpenBenchmarking.orgM samples/s, More Is BetterIndigoBench 4.0.64Scene: Supercar64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread246810SE +/- 0.20, N = 6SE +/- 0.04, N = 3SE +/- 0.01, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 36.565.673.161.740.970.510.25

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.0fRSA 4096-bit Performance64 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread13002600390052006500SE +/- 1.87, N = 3SE +/- 2.44, N = 3SE +/- 0.23, N = 3SE +/- 0.38, N = 3SE +/- 0.24, N = 3SE +/- 3.02, N = 3SE +/- 0.09, N = 35869.635514.603002.271582.67811.33422.73212.671. (CC) gcc options: -O3 -pthread -m64 -lssl -lcrypto -ldl

ebizzy

OpenBenchmarking.orgRecords/s, More Is Betterebizzy 0.364 Threads32 Threads16 Threads8 Threads4 Threads2 Threads1 Thread300K600K900K1200K1500KSE +/- 24098.90, N = 6SE +/- 22021.37, N = 6SE +/- 1118.92, N = 3SE +/- 738.51, N = 3SE +/- 282.86, N = 3SE +/- 3326.82, N = 3SE +/- 250.34, N = 3136771715554721145295671058343462177694909961. (CC) gcc options: -pthread -lpthread -O3 -march=native

Phoronix Test Suite v10.8.4