Linux Kernel Benchmarks AMD Opteron

AMD Opteron 2384 testing on Ubuntu 13.10 x86_64 with Linux 3.0, 3.4, 3.10, 3.12 kernels. Benchmarks by Michael Larabel for a future article on Phoronix.com.

HTML result view exported from: https://openbenchmarking.org/result/1311125-SO-LINUXKERN95.

Linux Kernel Benchmarks AMD OpteronProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionOpenGLLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0AMD Opteron 2384 @ 2.70GHz (4 Cores)TYAN S2927/S2927-ENVIDIA MCP554096MB64GB AGILITY-EXAMD Radeon HD 4870 512MBATI R6xx HDMIAcer P243WUbuntu 13.103.0.101-0300101-generic (x86_64)Xfce 4.10X Server 1.14.3radeon 7.2.0GCC 4.8ext41920x12003.4.68-030468-generic (x86_64)2.1 Mesa 9.2.1 Gallium 0.43.10.18-031018-generic (x86_64)3.1 Mesa 9.2.1 Gallium 0.43.12.0-031200-generic (x86_64)OpenBenchmarking.orgCompiler Details- --build=x86_64-linux-gnu --disable-browser-plugin --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,c++,java,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --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-ecj-jar=/usr/share/java/eclipse-ecj.jar --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-4.8-amd64/jre --with-jvm-jar-dir=/usr/lib/jvm-exports/java-1.5.0-gcj-4.8-amd64 --with-jvm-root-dir=/usr/lib/jvm/java-1.5.0-gcj-4.8-amd64 --with-multilib-list=m32,m64,mx32 --with-tune=generic -v Disk Details- Linux 3.0.101: CFQ / barrier=1,data=ordered,errors=remount-ro,relatime,rw,user_xattr- Linux 3.4.68: CFQ / data=ordered,errors=remount-ro,relatime,rw- Linux 3.10.18: DEADLINE / data=ordered,errors=remount-ro,relatime,rw- Linux 3.12.0: DEADLINE / data=ordered,errors=remount-ro,relatime,rwProcessor Details- Linux 3.0.101: Scaling Governor: powernow-k8 ondemand- Linux 3.4.68: Scaling Governor: powernow-k8 ondemand- Linux 3.10.18: Scaling Governor: acpi-cpufreq ondemand- Linux 3.12.0: Scaling Governor: acpi-cpufreq ondemandGraphics Details- EXASystem Details- Linux 3.0.101: Disk Scheduler: CFQ.- Linux 3.4.68: Disk Scheduler: CFQ.- Linux 3.10.18: Disk Scheduler: DEADLINE.- Linux 3.12.0: Disk Scheduler: DEADLINE.

Linux Kernel Benchmarks AMD Opteronfs-mark: 1000 Files, 1MB Sizefs-mark: 5000 Files, 1MB Size, 4 Threadsfs-mark: 4000 Files, 32 Sub Dirs, 1MB Sizepostmark: Disk Transaction Performancehpcc: G-HPLhpcc: G-Fftehpcc: EP-DGEMMhpcc: G-Ptranshpcc: EP-STREAM Triadhpcc: G-Rand Accesshpcc: Rand Ring Latencyhpcc: Rand Ring Bandwidthhpcc: Max Ping Pong Bandwidthparboil: Seven-Point Stencilparboil: Cutoff Pair Potentialparboil: Lid-Driven Cavity Fluid Dynamicsrodinia: OpenMP Leukocyterodinia: OpenMP CFD Solverrodinia: OpenMP Streamclusterhmmer: Pfam Database Searchmafft: Multiple Sequence Alignmenthimeno: Poisson Pressure Solverbuild-linux-kernel: Time To Compilec-ray: Total Timesmallpt: Global Illumination Renderer; 100 Samplesurbanterror: 1920 x 1200openarena: 1920 x 1200openarena: 1920 x 1200padman: 1920 x 1200xonotic: 1920 x 1200 - Lowxonotic: 1920 x 1200 - HighLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.021.2739.8221.45149712.252502.744375.583760.619062.368540.036770.716760.601142191.87093.4044.76544.2894.47237.4568.0623.3513.15538.48168.6054.4125020.9344.4022.68138612.421072.751755.864910.622811.806430.036990.734360.599342186.92992.5743.26544.3692.75234.1967.0223.0413.32510.65166.9554.0125021.55122.8058.47127.0362.5819.4720.9245.7321.25125812.429902.749606.148440.618071.887060.036760.726160.597292181.11693.0644.27543.2893.74236.3067.6822.8412.87504.28170.6854.0225161.93179.8069.33154.9769.9551.0021.1745.0722.57151812.452602.765716.205080.620061.802320.036800.720600.600932185.32592.8757.68543.1092.11234.3567.0223.5212.30540.63167.9054.0425064.20201.6071174.3089.4460.75OpenBenchmarking.org

FS-Mark

Test: 1000 Files, 1MB Size

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 1000 Files, 1MB SizeLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0510152025SE +/- 0.13, N = 3SE +/- 0.27, N = 3SE +/- 0.40, N = 6SE +/- 0.54, N = 621.2720.9320.9221.171. (CC) gcc options: -static

FS-Mark

Test: 5000 Files, 1MB Size, 4 Threads

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 5000 Files, 1MB Size, 4 ThreadsLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.01020304050SE +/- 2.18, N = 6SE +/- 2.42, N = 6SE +/- 1.52, N = 6SE +/- 1.72, N = 639.8244.4045.7345.071. (CC) gcc options: -static

FS-Mark

Test: 4000 Files, 32 Sub Dirs, 1MB Size

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 4000 Files, 32 Sub Dirs, 1MB SizeLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0510152025SE +/- 0.39, N = 6SE +/- 0.37, N = 4SE +/- 0.49, N = 6SE +/- 0.35, N = 321.4522.6821.2522.571. (CC) gcc options: -static

PostMark

Disk Transaction Performance

OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.030060090012001500SE +/- 15.59, N = 3SE +/- 2.33, N = 3SE +/- 13.04, N = 3SE +/- 7.94, N = 314971386125815181. (CC) gcc options: -O3

HPC Challenge

Test / Class: G-HPL

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: G-HPLLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.03691215SE +/- 0.05, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 312.2512.4212.4312.451. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: G-Ffte

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: G-FfteLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.00.62231.24461.86692.48923.1115SE +/- 0.00258, N = 3SE +/- 0.03836, N = 3SE +/- 0.00907, N = 3SE +/- 0.00405, N = 32.744372.751752.749602.765711. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: EP-DGEMM

OpenBenchmarking.orgGFLOPS, More Is BetterHPC Challenge 1.4.3Test / Class: EP-DGEMMLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0246810SE +/- 0.96527, N = 3SE +/- 0.72053, N = 3SE +/- 0.42035, N = 3SE +/- 0.42679, N = 35.583765.864916.148446.205081. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: G-Ptrans

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.4.3Test / Class: G-PtransLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.00.14010.28020.42030.56040.7005SE +/- 0.00903, N = 3SE +/- 0.00528, N = 3SE +/- 0.00604, N = 3SE +/- 0.00443, N = 30.619060.622810.618070.620061. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: EP-STREAM Triad

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.4.3Test / Class: EP-STREAM TriadLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.00.53291.06581.59872.13162.6645SE +/- 0.05860, N = 3SE +/- 0.00529, N = 3SE +/- 0.09207, N = 3SE +/- 0.00266, N = 32.368541.806431.887061.802321. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: G-Random Access

OpenBenchmarking.orgGUP/s, More Is BetterHPC Challenge 1.4.3Test / Class: G-Random AccessLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.00.00830.01660.02490.03320.0415SE +/- 0.00016, N = 3SE +/- 0.00075, N = 3SE +/- 0.00020, N = 3SE +/- 0.00041, N = 30.036770.036990.036760.036801. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: Random Ring Latency

OpenBenchmarking.orgusecs, Fewer Is BetterHPC Challenge 1.4.3Test / Class: Random Ring LatencyLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.00.16520.33040.49560.66080.826SE +/- 0.00618, N = 3SE +/- 0.00639, N = 3SE +/- 0.00902, N = 3SE +/- 0.00726, N = 30.716760.734360.726160.720601. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: Random Ring Bandwidth

OpenBenchmarking.orgGB/s, More Is BetterHPC Challenge 1.4.3Test / Class: Random Ring BandwidthLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.00.13530.27060.40590.54120.6765SE +/- 0.00133, N = 3SE +/- 0.00042, N = 3SE +/- 0.00093, N = 3SE +/- 0.00020, N = 30.601140.599340.597290.600931. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

HPC Challenge

Test / Class: Max Ping Pong Bandwidth

OpenBenchmarking.orgMB/s, More Is BetterHPC Challenge 1.4.3Test / Class: Max Ping Pong BandwidthLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.05001000150020002500SE +/- 10.83, N = 3SE +/- 5.01, N = 3SE +/- 9.63, N = 3SE +/- 6.00, N = 32191.872186.932181.122185.331. (CC) gcc options: -fomit-frame-pointer -O3 -march=native -funroll-loops -lblas -lm -pthread -lmpi -lopen-rte -lopen-pal -ldl -lnsl -lutil 2. BLAS + Open MPI 1.4.5

Parboil

Test: Seven-Point Stencil

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: Seven-Point StencilLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.020406080100SE +/- 0.26, N = 3SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.05, N = 393.4092.5793.0692.871. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: Cutoff Pair Potential

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: Cutoff Pair PotentialLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.01326395265SE +/- 0.68, N = 5SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.41, N = 344.7643.2644.2757.681. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Parboil

Test: Lid-Driven Cavity Fluid Dynamics

OpenBenchmarking.orgSeconds, Fewer Is BetterParboil 2.5Test: Lid-Driven Cavity Fluid DynamicsLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0120240360480600SE +/- 0.21, N = 3SE +/- 1.95, N = 3SE +/- 0.27, N = 3SE +/- 0.18, N = 3544.28544.36543.28543.101. (CXX) g++ options: -lm -lpthread -lgomp -ffast-math -fopenmp

Rodinia

Test: OpenMP Leukocyte

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP LeukocyteLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.020406080100SE +/- 0.18, N = 3SE +/- 0.14, N = 3SE +/- 0.23, N = 3SE +/- 0.07, N = 394.4792.7593.7492.111. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP CFD Solver

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP CFD SolverLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.050100150200250SE +/- 0.72, N = 3SE +/- 0.06, N = 3SE +/- 0.05, N = 3SE +/- 0.03, N = 3237.45234.19236.30234.351. (CXX) g++ options: -O2 -lOpenCL

Rodinia

Test: OpenMP Streamcluster

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 2.4Test: OpenMP StreamclusterLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.01530456075SE +/- 0.80, N = 3SE +/- 0.03, N = 3SE +/- 0.08, N = 3SE +/- 0.02, N = 368.0667.0267.6867.021. (CXX) g++ options: -O2 -lOpenCL

Timed HMMer Search

Pfam Database Search

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 2.3.2Pfam Database SearchLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0612182430SE +/- 0.28, N = 3SE +/- 0.19, N = 3SE +/- 0.04, N = 3SE +/- 0.46, N = 623.3523.0422.8423.521. (CC) gcc options: -O2 -pthread -lhmmer -lsquid -lm

Timed MAFFT Alignment

Multiple Sequence Alignment

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 6.864Multiple Sequence AlignmentLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.03691215SE +/- 0.20, N = 6SE +/- 0.03, N = 3SE +/- 0.30, N = 6SE +/- 0.29, N = 613.1513.3212.8712.301. (CC) gcc options: -O3 -lm -lpthread

Himeno Benchmark

Poisson Pressure Solver

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.0120240360480600SE +/- 1.54, N = 3SE +/- 2.36, N = 3SE +/- 1.94, N = 3SE +/- 3.33, N = 3538.48510.65504.28540.631. (CC) gcc options: -O3

Timed Linux Kernel Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 3.1Time To CompileLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.04080120160200SE +/- 1.74, N = 3SE +/- 1.21, N = 3SE +/- 1.19, N = 3SE +/- 1.11, N = 3168.60166.95170.68167.90

C-Ray

Total Time

OpenBenchmarking.orgSeconds, Fewer Is BetterC-Ray 1.1Total TimeLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.01224364860SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.03, N = 354.4154.0154.0254.041. (CC) gcc options: -lm -lpthread -O3

Smallpt

Global Illumination Renderer; 100 Samples

OpenBenchmarking.orgSeconds, Fewer Is BetterSmallpt 1.0Global Illumination Renderer; 100 SamplesLinux 3.0.101Linux 3.4.68Linux 3.10.18Linux 3.12.050100150200250SE +/- 0.88, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3SE +/- 0.00, N = 32502502512501. (CXX) g++ options: -fopenmp

Urban Terror

Resolution: 1920 x 1200

OpenBenchmarking.orgFrames Per Second, More Is BetterUrban Terror 4.2.013Resolution: 1920 x 1200Linux 3.4.68Linux 3.10.18Linux 3.12.01428425670SE +/- 7.69, N = 6SE +/- 0.22, N = 3SE +/- 0.21, N = 321.5561.9364.20

Urban Terror

Resolution: 1920 x 1200 - Total Frame Time

OpenBenchmarking.orgMilliseconds, Fewer Is BetterUrban Terror 4.2.013Resolution: 1920 x 1200 - Total Frame TimeLinux 3.4.68Linux 3.10.18Linux 3.12.020406080100Min: 2 / Avg: 18.69 / Max: 99Min: 2 / Avg: 16.06 / Max: 42Min: 2 / Avg: 15.46 / Max: 40

OpenArena

Resolution: 1920 x 1200

OpenBenchmarking.orgFrames Per Second, More Is BetterOpenArena 0.8.5Resolution: 1920 x 1200Linux 3.4.68Linux 3.10.18Linux 3.12.04080120160200SE +/- 0.70, N = 3SE +/- 0.42, N = 3SE +/- 1.07, N = 3122.80179.80201.60

OpenArena

Resolution: 1920 x 1200

OpenBenchmarking.orgFrames Per Second, More Is BetterOpenArena 0.8.8Resolution: 1920 x 1200Linux 3.4.68Linux 3.10.18Linux 3.12.01632486480SE +/- 0.07, N = 3SE +/- 0.41, N = 3SE +/- 0.00, N = 358.4769.3371.00MIN: 8MIN: 8MIN: 8

OpenArena

Resolution: 1920 x 1200 - Total Frame Time

OpenBenchmarking.orgMilliseconds, Fewer Is BetterOpenArena 0.8.8Resolution: 1920 x 1200 - Total Frame TimeLinux 3.4.68Linux 3.10.18Linux 3.12.01020304050Min: 8 / Avg: 17.07 / Max: 47Min: 8 / Avg: 14.37 / Max: 45Min: 8 / Avg: 14.05 / Max: 37

World of Padman

Resolution: 1920 x 1200

OpenBenchmarking.orgFrames Per Second, More Is BetterWorld of Padman 1.2Resolution: 1920 x 1200Linux 3.4.68Linux 3.10.18Linux 3.12.04080120160200SE +/- 0.52, N = 3SE +/- 0.28, N = 3SE +/- 0.31, N = 3127.03154.97174.30

Xonotic

Resolution: 1920 x 1200 - Effects Quality: Low

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.7Resolution: 1920 x 1200 - Effects Quality: LowLinux 3.4.68Linux 3.10.18Linux 3.12.020406080100SE +/- 0.22, N = 3SE +/- 0.03, N = 3SE +/- 0.11, N = 362.5869.9589.44MIN: 41 / MAX: 105MIN: 55 / MAX: 142MIN: 66 / MAX: 259

Xonotic

Resolution: 1920 x 1200 - Effects Quality: High

OpenBenchmarking.orgFrames Per Second, More Is BetterXonotic 0.7Resolution: 1920 x 1200 - Effects Quality: HighLinux 3.4.68Linux 3.10.18Linux 3.12.01428425670SE +/- 0.31, N = 6SE +/- 0.09, N = 3SE +/- 0.26, N = 319.4751.0060.75MIN: 6 / MAX: 89MIN: 25 / MAX: 85MIN: 23 / MAX: 127


Phoronix Test Suite v10.8.4