Linux 4.5 To Linux 4.7 P-State Scaling Governor Testing

Linux 4.5 to 4.7 kernel benchmarking. Tests by Michael Larabel.

HTML result view exported from: https://openbenchmarking.org/result/1605243-HA-LINUX47CP30&sor&grt.

ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionLinux 4.5Linux 4.6Linux 4.7 P-State Powersave P-State Performance P-State Powersave P-State Performance P-State Powersave P-State PerformanceIntel Xeon E5-2687W v3 @ 3.50GHz (20 Cores)MSI X99S SLI PLUS (MS-7885) v1.0Intel Xeon E7 v3/Xeon16384MBPNY CS1211 120GB + 80GB INTEL SSDSCKGW08AMD FirePro V7900 2048MBRealtek ALC892ASUS PB278Intel ConnectionUbuntu 16.044.5.0-040500-generic (x86_64)Unity 7.4.0X Server 1.18.3modesetting 1.18.34.1 Mesa 11.2.0 Gallium 0.4GCC 5.3.1 20160413ext42560x14404.6.0-040600-generic (x86_64)4.6.0-phx-schedutil (x86_64)OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-browser-plugin --disable-vtable-verify --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-gtk-cairo --enable-java-awt=gtk --enable-java-home --enable-languages=c,ada,c++,java,go,d,fortran,objc,obj-c++ --enable-libmpx --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-arch-directory=amd64 --with-default-libstdcxx-abi=new --with-multilib-list=m32,m64,mx32 --with-tune=generic -v Processor Details- Linux 4.5: P-State Powersave: Scaling Governor: intel_pstate powersave- Linux 4.5: P-State Performance: Scaling Governor: intel_pstate performance- Linux 4.6: P-State Powersave: Scaling Governor: intel_pstate powersave- Linux 4.6: P-State Performance: Scaling Governor: intel_pstate performance- Linux 4.7: P-State Powersave: Scaling Governor: intel_pstate powersave- Linux 4.7: P-State Performance: Scaling Governor: intel_pstate performanceGraphics Details- EXA

apache: Static Web Page Servingc-ray: Total Timeclomp: Static OMP Speedupffte: N=64, 1D Complex FFT Routineencode-flac: WAV To FLACgraphics-magick: Resizinghimeno: Poisson Pressure Solverhpcc: G-HPLjohn-the-ripper: Blowfishencode-mp3: WAV To MP3lammps: Rhodopsin Proteinmultichase: 256MB Array, 256 Byte Stridemultichase: 1GB Array, 256 Byte Stride, 4 Threadsnpb: EP.Bopm-git: Upscale-Relperm - 16pgbench: Buffer Test - Heavy Contention - Read Writeredis: GETbuild-linux-kernel: Time To Compilex264: H.264 Video EncodingLinux 4.5Linux 4.6Linux 4.7 P-State Powersave P-State Performance P-State Powersave P-State Performance P-State Powersave P-State Performance23233.6810.785.815254.797.081731765.48101.719001085712.7640.0468.4972.85356.9413.995281.93636557.3153.53351.9323433.6110.785.505286.287.081741787.37100.963001111812.9139.8767.3871.54346.5413.855367.98641185.4153.77352.4223123.5610.726.245281.597.071801782.62105.123671414112.7239.5966.7472.64370.6013.635311.59578049.2554.80350.8722942.7910.852.905109.4818.931711711.21101.876671012237.4240.6964.4774.17325.4114.065582.59569607.0896.57126.9013282.1629.576.111814.7018.9568625.0231.57813458437.29113.48107.03112.77111.974852.10387649.31136.33136.0213077.7629.596.424988.7418.9368623.92104.851671413637.41112.4469.4673.16105.234832.19387296.94135.89134.53OpenBenchmarking.org

Apache Benchmark

Static Web Page Serving

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page ServingP-State PerformanceP-State Powersave5K10K15K20K25KSE +/- 65.56, N = 3SE +/- 33.45, N = 3SE +/- 71.50, N = 3SE +/- 88.33, N = 3SE +/- 31.83, N = 3SE +/- 73.74, N = 323433.6123233.6823123.5622942.7913282.1613077.761. (CC) gcc options: -shared -fPIC -O2 -pthread

C-Ray

Total Time

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeP-State PowersaveP-State Performance714212835SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 3SE +/- 0.03, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 310.7210.8510.7810.7829.5729.591. (CC) gcc options: -lm -lpthread -O3

CLOMP

Static OMP Speedup

Linux 4.7Linux 4.6Linux 4.5OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 3.3Static OMP SpeedupP-State PerformanceP-State Powersave246810SE +/- 0.11, N = 5SE +/- 0.85, N = 5SE +/- 0.12, N = 10SE +/- 0.22, N = 10SE +/- 0.44, N = 10SE +/- 0.42, N = 106.426.116.242.905.815.501. (CC) gcc options: --openmp -O3 -lm

CPU Frequency (CPU0) Monitor

Phoronix Test Suite System Monitoring

OpenBenchmarking.orgMegahertzCPU Frequency (CPU0) MonitorPhoronix Test Suite System MonitoringLinux 4.7: P-State Performance6001200180024003000Min: 0 / Avg: 1512.39 / Max: 3500.93

FFTE

Test: N=64, 1D Complex FFT Routine

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 5.0Test: N=64, 1D Complex FFT RoutineP-State PerformanceP-State Powersave11002200330044005500SE +/- 1.60, N = 3SE +/- 27.12, N = 3SE +/- 5.83, N = 3SE +/- 11.10, N = 3SE +/- 146.87, N = 3SE +/- 0.06, N = 35286.285254.795281.595109.484988.741814.701. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi

FLAC Audio Encoding

WAV To FLAC

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.1WAV To FLACP-State PowersaveP-State Performance510152025SE +/- 0.06, N = 5SE +/- 0.01, N = 5SE +/- 0.07, N = 5SE +/- 0.05, N = 5SE +/- 0.01, N = 5SE +/- 0.01, N = 57.0718.937.087.0818.9318.951. (CXX) g++ options: -O2 -fvisibility=hidden -lm

GraphicsMagick

Operation: Resizing

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgIterations Per Minute, More Is BetterGraphicsMagick 1.3.19Operation: ResizingP-State PowersaveP-State Performance4080120160200SE +/- 0.00, N = 3SE +/- 0.58, N = 3SE +/- 1.20, N = 3SE +/- 0.58, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 318017117417368681. (CC) gcc options: -fopenmp -O2 -pthread -lXext -lSM -lICE -lX11 -lz -lm -lgomp -lpthread

Himeno Benchmark

Poisson Pressure Solver

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverP-State PerformanceP-State Powersave400800120016002000SE +/- 3.69, N = 3SE +/- 5.19, N = 3SE +/- 4.40, N = 3SE +/- 2.21, N = 3SE +/- 0.81, N = 3SE +/- 0.44, N = 31787.371765.481782.621711.21625.02623.921. (CC) gcc options: -O3 -mavx2

HPC Challenge

Test / Class: G-HPL

Linux 4.6Linux 4.7Linux 4.5OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: G-HPLP-State PowersaveP-State Performance20406080100SE +/- 0.03, N = 3SE +/- 0.21, N = 3SE +/- 0.03, N = 3SE +/- 0.40, N = 3SE +/- 0.10, N = 3SE +/- 0.07, N = 3105.12101.88104.8531.58101.72100.961. (CC) gcc options: -lblas -lm -pthread -lmpi -fomit-frame-pointer -O3 -march=native -funroll-loops2. BLAS + Open MPI 1.10.2

John The Ripper

Test: Blowfish

Linux 4.6Linux 4.7Linux 4.5OpenBenchmarking.orgReal C/S, More Is BetterJohn The Ripper 1.8.0Test: BlowfishP-State PowersaveP-State Performance3K6K9K12K15KSE +/- 9.67, N = 3SE +/- 34.07, N = 3SE +/- 0.00, N = 3SE +/- 411.91, N = 3SE +/- 564.91, N = 6SE +/- 504.45, N = 6141411012214136458411118108571. (CC) gcc options: -fopenmp -lcrypt

LAME MP3 Encoding

WAV To MP3

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.99.3WAV To MP3P-State PowersaveP-State Performance918273645SE +/- 0.08, N = 5SE +/- 0.32, N = 5SE +/- 0.06, N = 5SE +/- 0.17, N = 7SE +/- 0.21, N = 5SE +/- 0.12, N = 512.7237.4212.7612.9137.2937.411. (CC) gcc options: -O3 -ffast-math -funroll-loops -pipe -lncurses -lm

LAMMPS Molecular Dynamics Simulator

Test: Rhodopsin Protein

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgLoop Time, Fewer Is BetterLAMMPS Molecular Dynamics Simulator 1.0Test: Rhodopsin ProteinP-State PowersaveP-State Performance306090120150SE +/- 0.13, N = 3SE +/- 0.20, N = 3SE +/- 0.33, N = 3SE +/- 0.17, N = 3SE +/- 0.32, N = 3SE +/- 0.21, N = 339.5940.6939.8740.04112.44113.481. (CXX) g++ options: -lfftw -lmpich

Multichase Pointer Chaser

Test: 256MB Array, 256 Byte Stride

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgns, Fewer Is BetterMultichase Pointer ChaserTest: 256MB Array, 256 Byte StrideP-State PerformanceP-State Powersave20406080100SE +/- 1.02, N = 4SE +/- 0.38, N = 3SE +/- 1.05, N = 5SE +/- 0.68, N = 3SE +/- 0.05, N = 3SE +/- 0.27, N = 364.4766.7467.3868.4969.46107.031. (CC) gcc options: -O2 -static -pthread -lrt

Multichase Pointer Chaser

Test: 1GB Array, 256 Byte Stride, 4 Threads

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgns, Fewer Is BetterMultichase Pointer ChaserTest: 1GB Array, 256 Byte Stride, 4 ThreadsP-State PerformanceP-State Powersave306090120150SE +/- 0.51, N = 3SE +/- 0.99, N = 5SE +/- 0.99, N = 3SE +/- 0.57, N = 3SE +/- 0.64, N = 3SE +/- 0.12, N = 371.5472.8572.6474.1773.16112.771. (CC) gcc options: -O2 -static -pthread -lrt

NAS Parallel Benchmarks

Test / Class: EP.B

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.3Test / Class: EP.BP-State PowersaveP-State Performance80160240320400SE +/- 1.51, N = 3SE +/- 4.54, N = 6SE +/- 2.54, N = 3SE +/- 4.30, N = 3SE +/- 8.66, N = 3SE +/- 8.12, N = 3370.60325.41356.94346.54111.97105.231. (F9X) gfortran options: -fopenmp

Open Porous Media Git

OPM Benchmark: Upscale-Relperm - Threads: 16

Linux 4.6Linux 4.5OpenBenchmarking.orgSeconds, Fewer Is BetterOpen Porous Media GitOPM Benchmark: Upscale-Relperm - Threads: 16P-State PowersaveP-State Performance48121620SE +/- 0.20, N = 6SE +/- 0.01, N = 3SE +/- 0.21, N = 5SE +/- 0.18, N = 313.6314.0613.8513.991. Build Time Tue May 17 12:05:40 EDT 2016;

PostgreSQL pgbench

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

Linux 4.6Linux 4.5Linux 4.7OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 9.4.3Scaling: Buffer Test - Test: Heavy Contention - Mode: Read WriteP-State PerformanceP-State Powersave12002400360048006000SE +/- 2.84, N = 3SE +/- 146.77, N = 6SE +/- 75.30, N = 6SE +/- 90.16, N = 6SE +/- 9.49, N = 3SE +/- 4.75, N = 35582.595311.595367.985281.934852.104832.191. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -pthread -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

Redis

Test: GET

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 3.0.1Test: GETP-State PerformanceP-State Powersave140K280K420K560K700KSE +/- 10029.13, N = 4SE +/- 2523.79, N = 3SE +/- 8313.61, N = 3SE +/- 10015.82, N = 3SE +/- 674.73, N = 3SE +/- 229.02, N = 3641185.41636557.31578049.25569607.08387649.31387296.941. (CC) gcc options: -ggdb -rdynamic -lm -pthread -ldl -std=gnu99 -pipe -g3 -O3 -funroll-loops

Timed Linux Kernel Compilation

Time To Compile

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 4.3Time To CompileP-State PowersaveP-State Performance306090120150SE +/- 0.86, N = 3SE +/- 0.73, N = 3SE +/- 1.01, N = 3SE +/- 18.39, N = 6SE +/- 2.42, N = 3SE +/- 1.92, N = 353.5353.7754.8096.57135.89136.33

x264

H.264 Video Encoding

Linux 4.5Linux 4.6Linux 4.7OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2015-11-02H.264 Video EncodingP-State PerformanceP-State Powersave80160240320400SE +/- 4.32, N = 8SE +/- 4.34, N = 5SE +/- 5.10, N = 5SE +/- 0.95, N = 5SE +/- 0.57, N = 5SE +/- 0.56, N = 5352.42351.93350.87126.90136.02134.531. (CC) gcc options: -ldl -m64 -lm -lpthread -O3 -ffast-math -std=gnu99 -fomit-frame-pointer -fno-tree-vectorize

Phoronix Test Suite v10.8.5