CompuLab Utilite Quad-Core ARM vs. Fit-PC2, TrimSlice

Testing by Michael Larabel of Phoronix for a future article of the different CompuLab devices... More benchmarks coming this is just a teaser.

HTML result view exported from: https://openbenchmarking.org/result/1401254-PL-TRIMSLICE07&sor&grs.

CompuLab Utilite Quad-Core ARM vs. Fit-PC2, TrimSliceProcessorMotherboardMemoryDiskGraphicsNetworkChipsetAudioMonitorOSKernelDesktopDisplay ServerOpenGLCompilerFile-SystemScreen ResolutionDisplay DriverCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-SliceARMv7 rev 10 @ 1.00GHz (4 Cores)Compulab CM-FX62048MB32GB SanDisk SSD U100GC2000 EngineIntel I211 Gigabit ConnectionUbuntu 12.043.0.35-cm-fx6-5.1 (armv7l)GNOME 3.2.1X Server 1.11.32.1GCC 4.6ext41920x1080Intel Atom Z530 @ 1.60GHz (2 Cores)Intel SBC-FITPC2Intel Hub + SCH1024MB160GB Hitachi HTS54501Intel Hub (SCH Poulsbo)Realtek ALC260DELL S2409WRealtek RTL8111/8168B + Ralink RT3090 Wireless 802.11n 1T/1R3.8.0-29-generic (i686)modesetting 0.7.0ARMv7 rev 0 @ 1.00GHz (2 Cores)trimslice593MB250GB Samsung HM251HINVIDIA TEGRARealtek RTL8111/8168B3.1.10-l4t.r16.01 (armv7l)ext21366x1536OpenBenchmarking.orgCompiler Details- CompuLab Utilite: --build=arm-linux-gnueabi --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabi --target=arm-linux-gnueabi --with-arch=armv7-a --with-float=softfp --with-fpu=vfpv3-d16 --with-mode=thumb -v - CompuLab Fit-PC2: --build=i686-linux-gnu --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-targets=all --enable-threads=posix --host=i686-linux-gnu --target=i686-linux-gnu --with-arch-32=i686 --with-tune=generic -v - CompuLab Trim-Slice: --build=arm-linux-gnueabihf --disable-sjlj-exceptions --disable-werror --enable-checking=release --enable-clocale=gnu --enable-gnu-unique-object --enable-languages=c,c++,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multilib --enable-nls --enable-objc-gc --enable-plugin --enable-shared --enable-threads=posix --host=arm-linux-gnueabihf --target=arm-linux-gnueabihf --with-arch=armv7-a --with-float=hard --with-fpu=vfpv3-d16 --with-mode=thumb -v Processor Details- CompuLab Utilite: Scaling Governor: imx ondemand- CompuLab Fit-PC2: Scaling Governor: acpi-cpufreq ondemand- CompuLab Trim-Slice: Scaling Governor: tegra ondemand

CompuLab Utilite Quad-Core ARM vs. Fit-PC2, TrimSlicehpcc: G-Rand Accessscimark2: Sparse Matrix Multiplyc-ray: Total Timescimark2: Dense LU Matrix Factorizationhpcc: G-HPLscimark2: Jacobi Successive Over-Relaxationsmallpt: Global Illumination Renderer; 100 Samplesn-queens: Elapsed Timescimark2: Compositehpcc: Rand Ring Latencyhpcc: EP-STREAM Triadstream: Scaledolfyn: Computational Fluid Dynamicsstream: Triadstream: Addcompress-gzip: 2GB File Compressionstream: Copyhpcc: G-Fftetscp: AI Chess Performancehpcc: G-Ptranspolybench-c: Correlation Computationpolybench-c: Covariance Computationbuild-apache: Time To Compileffmpeg: H.264 HD To NTSC DVffte: N=64, 1D Complex FFT Routinehpcc: EP-DGEMMpolybench-c: 3 Matrix Multiplicationsscimark2: Monte Carloencode-flac: WAV To FLACscimark2: Fast Fourier Transformhimeno: Poisson Pressure Solverencode-mp3: WAV To MP3apache: Static Web Page Servingopenssl: RSA 4096-bit Performancex264: H.264 Video Encodingcrafty: Elapsed Timeprimesieve: 1e12 Prime Number Generationclomp: Static OMP Speeduphpcc: Max Ping Pong Bandwidthhpcc: Rand Ring BandwidthCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice0.0016252.73630.4767.532.01655124.121419246.5962.154.328370.514841438.23578.601652.731678.17102.701486.550.249771279790.0897124.5924.53390.00357.63387.340.56971571.4849.8056.8216.57109.12128.792512.2416.609.060.111382.414.63711.9560.141960.00043156.461823.45153.000.82762270.602559128.868.758240.850121506.41325.911672.491670.9786.341447.700.187201979360.1140431.2230.88473.04287.680.44101485.7748.1156.7016.15104.05135.490.032264.13521.5110.222990.0006954.771261.4553.981.04644140.103049518.4464.535.162310.44687846.85551.51977.26985.78146.01890.480.156941264510.1319921.9721.87537.40489.530.56679551.9356.3361.4417.48103.47135.750.232513.161.931280.5530.58702OpenBenchmarking.org

HPC Challenge

Test / Class: G-Random Access

OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.4.3Test / Class: G-Random AccessCompuLab UtiliteCompuLab Trim-SliceCompuLab Fit-PC20.00040.00080.00120.00160.002SE +/- 0.00001, N = 3SE +/- 0.00000, N = 3SE +/- 0.00000, N = 30.001620.000690.000431. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

SciMark

Computational Test: Sparse Matrix Multiply

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Sparse Matrix MultiplyCompuLab Fit-PC2CompuLab Trim-SliceCompuLab Utilite306090120150SE +/- 0.23, N = 4SE +/- 0.41, N = 4SE +/- 0.05, N = 4156.4654.7752.73

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeCompuLab UtiliteCompuLab Trim-SliceCompuLab Fit-PC2400800120016002000SE +/- 4.34, N = 3SE +/- 0.24, N = 3SE +/- 0.19, N = 3630.471261.451823.451. (CC) gcc options: -lm -lpthread -O3

SciMark

Computational Test: Dense LU Matrix Factorization

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Dense LU Matrix FactorizationCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice306090120150SE +/- 0.09, N = 4SE +/- 0.06, N = 4SE +/- 0.29, N = 4153.0067.5353.98

HPC Challenge

Test / Class: G-HPL

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: G-HPLCompuLab UtiliteCompuLab Trim-SliceCompuLab Fit-PC20.45370.90741.36111.81482.2685SE +/- 0.00153, N = 3SE +/- 0.00605, N = 3SE +/- 0.00373, N = 32.016551.046440.827621. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

SciMark

Computational Test: Jacobi Successive Over-Relaxation

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Jacobi Successive Over-RelaxationCompuLab Fit-PC2CompuLab Trim-SliceCompuLab Utilite60120180240300SE +/- 0.24, N = 4SE +/- 3.31, N = 4SE +/- 0.10, N = 4270.60140.10124.12

Smallpt

Global Illumination Renderer; 100 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice7001400210028003500SE +/- 0.58, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 31419255930491. (CXX) g++ options: -fopenmp

N-Queens

Elapsed Time

OpenBenchmarking.orgSeconds, Fewer Is BetterN-Queens 1.0Elapsed TimeCompuLab UtiliteCompuLab Trim-Slice110220330440550SE +/- 4.80, N = 3SE +/- 0.07, N = 3246.59518.441. (CC) gcc options: -static -fopenmp -O3

SciMark

Computational Test: Composite

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: CompositeCompuLab Fit-PC2CompuLab Trim-SliceCompuLab Utilite306090120150SE +/- 0.06, N = 4SE +/- 0.53, N = 4SE +/- 0.07, N = 4128.8664.5362.15

HPC Challenge

Test / Class: Random Ring Latency

OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.4.3Test / Class: Random Ring LatencyCompuLab UtiliteCompuLab Trim-SliceCompuLab Fit-PC2246810SE +/- 0.02778, N = 3SE +/- 0.01181, N = 3SE +/- 0.05629, N = 34.328375.162318.758241. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

HPC Challenge

Test / Class: EP-STREAM Triad

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.4.3Test / Class: EP-STREAM TriadCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice0.19130.38260.57390.76520.9565SE +/- 0.00628, N = 3SE +/- 0.00478, N = 3SE +/- 0.00207, N = 30.850120.514840.446871. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

Stream

Type: Scale

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: ScaleCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice30060090012001500SE +/- 0.90, N = 9SE +/- 2.72, N = 8SE +/- 0.56, N = 81506.411438.23846.851. (CC) gcc options: -O3 -march=native -fopenmp

Dolfyn

Computational Fluid Dynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterDolfyn 0.527Computational Fluid DynamicsCompuLab Fit-PC2CompuLab Trim-SliceCompuLab Utilite130260390520650SE +/- 0.34, N = 3SE +/- 2.88, N = 3SE +/- 1.29, N = 3325.91551.51578.60

Stream

Type: Triad

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: TriadCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice400800120016002000SE +/- 1.19, N = 7SE +/- 3.10, N = 10SE +/- 0.63, N = 101672.491652.73977.261. (CC) gcc options: -O3 -march=native -fopenmp

Stream

Type: Add

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: AddCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice400800120016002000SE +/- 2.89, N = 10SE +/- 0.97, N = 10SE +/- 0.63, N = 91678.171670.97985.781. (CC) gcc options: -O3 -march=native -fopenmp

Gzip Compression

2GB File Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterGzip Compression2GB File CompressionCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice306090120150SE +/- 0.42, N = 3SE +/- 0.14, N = 3SE +/- 0.62, N = 386.34102.70146.01

Stream

Type: Copy

OpenBenchmarking.orgMB/s, More Is BetterStream 2013-01-17Type: CopyCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice30060090012001500SE +/- 2.65, N = 10SE +/- 1.48, N = 10SE +/- 2.04, N = 101486.551447.70890.481. (CC) gcc options: -O3 -march=native -fopenmp

HPC Challenge

Test / Class: G-Ffte

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: G-FfteCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice0.05620.11240.16860.22480.281SE +/- 0.00089, N = 3SE +/- 0.00030, N = 3SE +/- 0.00186, N = 30.249770.187200.156941. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

TSCP

AI Chess Performance

OpenBenchmarking.orgNodes Per Second, More Is BetterTSCP 1.81AI Chess PerformanceCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice40K80K120K160K200KSE +/- 42.80, N = 5SE +/- 843.03, N = 5SE +/- 1100.08, N = 51979361279791264511. (CC) gcc options: -O3 -march=native

HPC Challenge

Test / Class: G-Ptrans

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.4.3Test / Class: G-PtransCompuLab Trim-SliceCompuLab Fit-PC2CompuLab Utilite0.02970.05940.08910.11880.1485SE +/- 0.00197, N = 3SE +/- 0.00025, N = 3SE +/- 0.00203, N = 30.131990.114040.089711. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

PolyBench-C

Test: Correlation Computation

OpenBenchmarking.orgSeconds, Fewer Is BetterPolyBench-C 3.2Test: Correlation ComputationCompuLab Trim-SliceCompuLab UtiliteCompuLab Fit-PC2714212835SE +/- 0.21, N = 3SE +/- 0.01, N = 3SE +/- 0.46, N = 621.9724.5931.221. (CC) gcc options: -O3

PolyBench-C

Test: Covariance Computation

OpenBenchmarking.orgSeconds, Fewer Is BetterPolyBench-C 3.2Test: Covariance ComputationCompuLab Trim-SliceCompuLab UtiliteCompuLab Fit-PC2714212835SE +/- 0.19, N = 3SE +/- 0.08, N = 3SE +/- 0.24, N = 321.8724.5330.881. (CC) gcc options: -O3

Timed Apache Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.7Time To CompileCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice120240360480600SE +/- 0.79, N = 3SE +/- 0.10, N = 3SE +/- 0.33, N = 3390.00473.04537.40

FFmpeg

H.264 HD To NTSC DV

OpenBenchmarking.orgSeconds, Fewer Is BetterFFmpeg 2.1.1H.264 HD To NTSC DVCompuLab UtiliteCompuLab Trim-Slice110220330440550SE +/- 1.40, N = 3SE +/- 6.63, N = 3357.63489.531. (CC) gcc options: -lavdevice -lavfilter -lavformat -lavcodec -lswresample -lswscale -lavutil -ldl -lm -pthread -lrt -march=armv7-a -std=c99 -fomit-frame-pointer -mthumb -O3 -fno-math-errno -fno-signed-zeros -fno-tree-vectorize -MMD -MF -MT

FFTE

Test: N=64, 1D Complex FFT Routine

OpenBenchmarking.orgMFLOPS, More Is BetterFFTE 5.0Test: N=64, 1D Complex FFT RoutineCompuLab UtiliteCompuLab Fit-PC280160240320400SE +/- 0.47, N = 3SE +/- 0.06, N = 3387.34287.681. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -lmpichf90 -lmpich -lopa -lmpl -lrt -lcr -lpthread

HPC Challenge

Test / Class: EP-DGEMM

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: EP-DGEMMCompuLab UtiliteCompuLab Trim-SliceCompuLab Fit-PC20.12820.25640.38460.51280.641SE +/- 0.00029, N = 3SE +/- 0.00877, N = 3SE +/- 0.01115, N = 30.569710.566790.441011. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

PolyBench-C

Test: 3 Matrix Multiplications

OpenBenchmarking.orgSeconds, Fewer Is BetterPolyBench-C 3.2Test: 3 Matrix MultiplicationsCompuLab Fit-PC2CompuLab Trim-SliceCompuLab Utilite120240360480600SE +/- 0.41, N = 3SE +/- 0.45, N = 3SE +/- 0.96, N = 3485.77551.93571.481. (CC) gcc options: -O3

SciMark

Computational Test: Monte Carlo

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Monte CarloCompuLab Trim-SliceCompuLab UtiliteCompuLab Fit-PC21326395265SE +/- 0.95, N = 4SE +/- 0.09, N = 4SE +/- 0.00, N = 456.3349.8048.11

FLAC Audio Encoding

WAV To FLAC

OpenBenchmarking.orgSeconds, Fewer Is BetterFLAC Audio Encoding 1.3.0WAV To FLACCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice1428425670SE +/- 0.06, N = 5SE +/- 0.20, N = 5SE +/- 0.32, N = 556.7056.8261.441. (CXX) g++ options: -O2 -fvisibility=hidden -lm

SciMark

Computational Test: Fast Fourier Transform

OpenBenchmarking.orgMflops, More Is BetterSciMark 2.0Computational Test: Fast Fourier TransformCompuLab Trim-SliceCompuLab UtiliteCompuLab Fit-PC248121620SE +/- 0.20, N = 4SE +/- 0.07, N = 4SE +/- 0.01, N = 417.4816.5716.15

Himeno Benchmark

Poisson Pressure Solver

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice20406080100SE +/- 1.14, N = 3SE +/- 0.35, N = 3SE +/- 0.62, N = 3109.12104.05103.471. (CC) gcc options: -O3

LAME MP3 Encoding

WAV To MP3

OpenBenchmarking.orgSeconds, Fewer Is BetterLAME MP3 Encoding 3.99.3WAV To MP3CompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice306090120150SE +/- 0.52, N = 5SE +/- 0.08, N = 5SE +/- 0.09, N = 5128.79135.49135.751. (CC) gcc options: -O3 -fomit-frame-pointer -ffast-math -pipe -lm

Apache Benchmark

Static Web Page Serving

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page ServingCompuLab Utilite5001000150020002500SE +/- 1.78, N = 32512.241. (CC) gcc options: -shared -fPIC -O2 -pthread

OpenSSL

RSA 4096-bit Performance

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.0.1fRSA 4096-bit PerformanceCompuLab Utilite48121620SE +/- 0.00, N = 316.601. (CC) gcc options: -march=armv7-a -O3 -lssl -lcrypto -ldl

x264

H.264 Video Encoding

OpenBenchmarking.orgFrames Per Second, More Is Betterx264 2014-01-09H.264 Video EncodingCompuLab Utilite3691215SE +/- 0.08, N = 59.061. (CC) gcc options: -ldl -lm -lpthread

Crafty

Elapsed Time

OpenBenchmarking.orgSeconds, Fewer Is BetterCrafty 23.4Elapsed TimeCompuLab Fit-PC2CompuLab UtiliteCompuLab Trim-Slice0.05180.10360.15540.20720.259SE +/- 0.00, N = 6SE +/- 0.00, N = 6SE +/- 0.03, N = 60.030.110.231. (CC) gcc options: -lstdc++ -lm

Primesieve

1e12 Prime Number Generation

OpenBenchmarking.orgSeconds, Fewer Is BetterPrimesieve 4.21e12 Prime Number GenerationCompuLab UtiliteCompuLab Fit-PC2CompuLab Trim-Slice5001000150020002500SE +/- 44.53, N = 3SE +/- 3.22, N = 3SE +/- 258.63, N = 31382.412264.132513.161. (CXX) g++ options: -O2 -fopenmp

CLOMP

Static OMP Speedup

OpenBenchmarking.orgSpeedup, More Is BetterCLOMP 3.3Static OMP SpeedupCompuLab UtiliteCompuLab Trim-Slice1.04182.08363.12544.16725.209SE +/- 0.27, N = 10SE +/- 0.01, N = 54.631.931. (CC) gcc options: --openmp -O3 -lm

HPC Challenge

Test / Class: Max Ping Pong Bandwidth

OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.4.3Test / Class: Max Ping Pong BandwidthCompuLab Trim-SliceCompuLab UtiliteCompuLab Fit-PC230060090012001500SE +/- 5.35, N = 3SE +/- 7.87, N = 3SE +/- 29.05, N = 31280.55711.96521.511. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

HPC Challenge

Test / Class: Random Ring Bandwidth

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.4.3Test / Class: Random Ring BandwidthCompuLab Trim-SliceCompuLab Fit-PC2CompuLab Utilite0.13210.26420.39630.52840.6605SE +/- 0.00207, N = 3SE +/- 0.01334, N = 3SE +/- 0.00126, N = 30.587020.222990.141961. (CC) gcc options: -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil -fomit-frame-pointer -O3 -march=native -funroll-loops 2. BLAS + Open MPI 1.4.3

Phoronix Test Suite v10.8.4