POWER9 Spectre mitigation benchmarks by Michael Larabel.
Kernel Protection Compiler Notes: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -vProcessor Notes: Scaling Governor: powernv-cpufreq ondemandPython Notes: Python 2.7.15+ + Python 3.6.7Security Notes: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch serialisation (kernel only) + Kernel entry/exit barrier (eieio)
Default Kernel + User Protection Compiler Notes: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: powernv-cpufreq ondemandPython Notes: Python 2.7.15+ + Python 3.6.7Security Notes: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch cache disabled + Kernel entry/exit barrier (eieio)
No Protection Processor: POWER9 altivec supported @ 3.80GHz (44 Cores / 176 Threads), Motherboard: PowerNV T2P9D01 REV 1.01, Memory: 65536MB, Disk: Samsung SSD 960 EVO 500GB, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom NetXtreme BCM5719 Gigabit PCIe
OS: Ubuntu 18.10, Kernel: 4.18.0-10-generic (ppc64le), Compiler: GCC 8.2.0, File-System: ext4, Screen Resolution: 1920x1080
Compiler Notes: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: powernv-cpufreq ondemandPython Notes: Python 2.7.15+ + Python 3.6.7
OldFart Fedora Processor: AMD A6-3410MX APU @ 1.60GHz (4 Cores) , Motherboard: HP 358D v33.18 (F.21 BIOS) , Chipset: AMD Family 12h Root Complex , Memory: 6144MB , Disk: 1000GB Samsung SSD 840 , Graphics: AMD Radeon HD 6520G 512MB , Audio: AMD BeaverCreek HDMI Audio , Network: Realtek RTL8111/8168/8411 + Broadcom and subsidiaries BCM4313 802.11bgn Wireless
OS: Fedora 29, Kernel: 4.19.1jan (x86_64), Desktop: GNOME Shell 3.30.1, Display Server: Wayland, OpenGL: 3.3 Mesa 18.2.4 (LLVM 7.0.0), File-System: ext4, Screen Resolution: 1366x768
Disk Notes: BFQ / relatime,rw,seclabelProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15 + Python 3.7.1Security Notes: SELinux + __user pointer sanitization + Full AMD retpoline
Talos II Dual 22-Core POWER9 Spectre Benchmarks Processor Motherboard Memory Disk Graphics Monitor Network Chipset Audio OS Kernel Compiler File-System Screen Resolution Desktop Display Server OpenGL Kernel Protection Default Kernel + User Protection No Protection OldFart Fedora POWER9 altivec supported @ 3.80GHz (44 Cores / 176 Threads) PowerNV T2P9D01 REV 1.01 65536MB Samsung SSD 960 EVO 500GB ASPEED ASPEED Family VE228 Broadcom NetXtreme BCM5719 Gigabit PCIe Ubuntu 18.10 4.18.0-10-generic (ppc64le) GCC 8.2.0 ext4 1920x1080 AMD A6-3410MX APU @ 1.60GHz (4 Cores) HP 358D v33.18 (F.21 BIOS) AMD Family 12h Root Complex 6144MB 1000GB Samsung SSD 840 AMD Radeon HD 6520G 512MB AMD BeaverCreek HDMI Audio Realtek RTL8111/8168/8411 + Broadcom and subsidiaries BCM4313 802.11bgn Wireless Fedora 29 4.19.1jan (x86_64) GNOME Shell 3.30.1 Wayland 3.3 Mesa 18.2.4 (LLVM 7.0.0) 1366x768 OpenBenchmarking.org Compiler Details - Kernel Protection, Default Kernel + User Protection, No Protection: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -v Processor Details - Kernel Protection: Scaling Governor: powernv-cpufreq ondemand - Default Kernel + User Protection: Scaling Governor: powernv-cpufreq ondemand - No Protection: Scaling Governor: powernv-cpufreq ondemand - OldFart Fedora: Scaling Governor: acpi-cpufreq ondemand Python Details - Kernel Protection: Python 2.7.15+ + Python 3.6.7 - Default Kernel + User Protection: Python 2.7.15+ + Python 3.6.7 - No Protection: Python 2.7.15+ + Python 3.6.7 - OldFart Fedora: Python 2.7.15 + Python 3.7.1 Security Details - Kernel Protection: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch serialisation (kernel only) + Kernel entry/exit barrier (eieio) - Default Kernel + User Protection: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch cache disabled + Kernel entry/exit barrier (eieio) - OldFart Fedora: SELinux + __user pointer sanitization + Full AMD retpoline Disk Details - Default Kernel + User Protection: NONE / errors=remount-ro,relatime,rw - No Protection: NONE / errors=remount-ro,relatime,rw - OldFart Fedora: BFQ / relatime,rw,seclabel
Kernel Protection Default Kernel + User Protection No Protection OldFart Fedora Logarithmic Result Overview Phoronix Test Suite Timed ImageMagick Compilation Cython benchmark Compile Bench C-Ray OpenSSL Stockfish Hackbench POV-Ray Stress-NG Timed LLVM Compilation PyBench BYTE Unix Benchmark AOBench PHPBench Timed Linux Kernel Compilation Primesieve Rodinia Rust Prime Benchmark GNU MPC Go Benchmarks Bork File Encrypter
Talos II Dual 22-Core POWER9 Spectre Benchmarks build-linux-kernel: Time To Compile hackbench: 32 - Process build-llvm: Time To Compile c-ray: Total Time - 4K, 16 Rays Per Pixel stockfish: Total Time povray: Trace Time rodinia: OpenMP LavaMD byte: Dhrystone 2 cython-bench: rust-prime: Prime Number Test To 200,000,000 build-imagemagick: Time To Compile rodinia: OpenMP CFD Solver aobench: 2048 x 2048 - Total Time stress-ng: Hsearch pybench: Total For Average Test Times primesieve: 1e12 Prime Number Generation mpcbench: Multi-Precision Benchmark go-benchmark: build stress-ng: Glibc Qsort Data Sorting stress-ng: Lsearch stress-ng: System V Message Passing compilebench: Compile rodinia: OpenMP Streamcluster phpbench: PHP Benchmark Suite bork: File Encryption Time go-benchmark: http stress-ng: Semaphores stress-ng: Memory Copying stress-ng: CPU Stress stress-ng: Tsearch stress-ng: Forking stress-ng: Socket Activity stress-ng: Crypto stress-ng: Vector Math stress-ng: Glibc C String Functions go-benchmark: garbage openssl: RSA 4096-bit Performance go-benchmark: json ffte: N=256, 1D Complex FFT Routine compilebench: Read Compiled Tree compilebench: Initial Create Kernel Protection Default Kernel + User Protection No Protection OldFart Fedora 50.66 27.17 174 19.05 79799227 24.92 18.37 26258513 40.66 3.72 27.77 13.25 59.51 186931 1939 17.49 5370 28049160231 368 386 6492204 2533 29.60 330044 22.22 1928510 347166 8340 12321 889 58592 27094 8149 88416 1370824 1211245 7417 4238819 6770 1823 346 55.46 28.14 185 18.93 79135103 25.99 18.76 26755821 66.62 3.51 30.03 14.79 64.61 164138 4090 17.47 4007 29899607203 253 202 6293617 2324 29.77 159731 22.51 1341717 347517 8317 12279 859 59360 26272 8075 88154 1290587 1233700 7394 4898644 1538 296 50.01 23.89 172 17.83 83011846 21.85 16.75 26618718 38.88 3.32 27.28 12.93 59.52 201660 1816 16.38 5433 27993756394 369 389 6607012 2539 28.57 329783 22.16 1568381 456638 8368 12402 898 60585 28717 8298 88699 1379226 1106825 7390 3813471 1887 348 747.33 715.57 3668 697.31 2680116 560.58 464.11 11840121 289.35 398.06 235.73 290.45 131.35 4748 4987 232.79 2827 59019624561 11.98 13.70 437650 427 109.13 168448 36.26 54528 1694586 203 284 30.95 4864 308 237 1747 31256 12861996 200 57034155 315 96.59 OpenBenchmarking.org
Hackbench This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 32 - Type: Process No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 150 300 450 600 750 SE +/- 0.04, N = 3 SE +/- 0.21, N = 3 SE +/- 0.02, N = 3 SE +/- 11.26, N = 9 23.89 27.17 28.14 715.57 1. (CC) gcc options: -lpthread
C-Ray This is a test of C-Ray, a simple raytracer designed to test the floating-point CPU performance. This test is multi-threaded (16 threads per core), will shoot 8 rays per pixel for anti-aliasing, and will generate a 1600 x 1200 image. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel No Protection Default Kernel + User Protection Kernel Protection OldFart Fedora 150 300 450 600 750 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 SE +/- 0.20, N = 3 17.83 18.93 19.05 697.31 1. (CC) gcc options: -lm -lpthread -O3
Stockfish This is a test of Stockfish, an advanced C++11 chess benchmark that can scale up to 128 CPU cores. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nodes Per Second, More Is Better Stockfish 9 Total Time No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 20M 40M 60M 80M 100M SE +/- 743288.42, N = 3 SE +/- 1474158.28, N = 3 SE +/- 662670.61, N = 3 SE +/- 42825.92, N = 3 83011846 79799227 79135103 2680116 1. (CXX) g++ options: -m64 -lpthread -fno-exceptions -std=c++11 -pedantic -O3 -flto
POV-Ray This is a test of POV-Ray, the Persistence of Vision Raytracer. POV-Ray is used to create 3D graphics using ray-tracing. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better POV-Ray 3.7.0.7 Trace Time No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 120 240 360 480 600 SE +/- 0.15, N = 3 SE +/- 0.13, N = 3 SE +/- 0.09, N = 3 SE +/- 7.80, N = 3 21.85 24.92 25.99 560.58 -march=native 1. (CXX) g++ options: -pipe -O3 -ffast-math -pthread -ltiff -ljpeg -lpng -lz -lrt -lm -lboost_thread -lboost_system
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 100 200 300 400 500 SE +/- 0.20, N = 3 SE +/- 0.31, N = 4 SE +/- 0.26, N = 3 SE +/- 6.42, N = 3 16.75 18.37 18.76 464.11 1. (CXX) g++ options: -O2 -lOpenCL
Rust Prime Benchmark Based on petehunt/rust-benchmark, this is a prime number benchmark that is multi-threaded and written in Rustlang. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rust Prime Benchmark Prime Number Test To 200,000,000 No Protection Default Kernel + User Protection Kernel Protection OldFart Fedora 90 180 270 360 450 SE +/- 0.06, N = 3 SE +/- 0.06, N = 3 SE +/- 0.07, N = 3 SE +/- 1.15, N = 3 3.32 3.51 3.72 398.06 -ldl -lrt -lpthread -lgcc_s -lc -lm -lutil 1. (CC) gcc options: -m64 -pie -nodefaultlibs
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 60 120 180 240 300 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.08, N = 3 SE +/- 2.82, N = 3 12.93 13.25 14.79 290.45 1. (CXX) g++ options: -O2 -lOpenCL
AOBench AOBench is a lightweight ambient occlusion renderer, written in C. The test profile is using a size of 2048 x 2048. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better AOBench Size: 2048 x 2048 - Total Time Kernel Protection No Protection Default Kernel + User Protection OldFart Fedora 30 60 90 120 150 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.32, N = 3 59.51 59.52 64.61 131.35 1. (CC) gcc options: -lm -O3
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Hsearch No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 40K 80K 120K 160K 200K SE +/- 1283.68, N = 3 SE +/- 4467.03, N = 12 SE +/- 3274.73, N = 12 SE +/- 5.86, N = 3 201660 186931 164138 4748 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
PyBench This test profile reports the total time of the different average timed test results from PyBench. PyBench reports average test times for different functions such as BuiltinFunctionCalls and NestedForLoops, with this total result providing a rough estimate as to Python's average performance on a given system. This test profile runs PyBench each time for 20 rounds. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 1100 2200 3300 4400 5500 SE +/- 3.06, N = 3 SE +/- 5.24, N = 3 SE +/- 0.88, N = 3 1816 1939 4090 4987
Primesieve Primesieve generates prime numbers using a highly optimized sieve of Eratosthenes implementation. Primesieve benchmarks the CPU's L1/L2 cache performance. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Primesieve 7.1 1e12 Prime Number Generation No Protection Default Kernel + User Protection Kernel Protection OldFart Fedora 50 100 150 200 250 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.03, N = 3 SE +/- 1.32, N = 3 16.38 17.47 17.49 232.79 -O2 1. (CXX) g++ options: -lpthread
GNU MPC GNU MPC is a C library for the arithmetic of complex numbers. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Global Score, More Is Better GNU MPC 1.1.0 Multi-Precision Benchmark No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 1200 2400 3600 4800 6000 SE +/- 3.33, N = 3 SE +/- 3.33, N = 3 SE +/- 3.33, N = 3 5433 5370 4007 2827 -mtune=power8 -O3 -mtune=power8 -O3 -mtune=power8 -O3 -lm -O2 -pedantic -fomit-frame-pointer -mtune=amdfam10 -march=amdfam10 1. (CC) gcc options: -m64 -MT -MD -MP -MF
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: build No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 13000M 26000M 39000M 52000M 65000M SE +/- 318891587.38, N = 12 SE +/- 94097418.28, N = 3 SE +/- 573457278.59, N = 3 SE +/- 722329537.55, N = 3 27993756394 28049160231 29899607203 59019624561
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Glibc Qsort Data Sorting No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 80 160 240 320 400 SE +/- 1.48, N = 3 SE +/- 0.71, N = 3 SE +/- 0.49, N = 3 SE +/- 0.22, N = 12 369.00 368.00 253.00 11.98 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Lsearch No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 80 160 240 320 400 SE +/- 0.70, N = 3 SE +/- 1.09, N = 3 SE +/- 1.34, N = 3 SE +/- 0.31, N = 12 389.00 386.00 202.00 13.70 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: System V Message Passing No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 1.4M 2.8M 4.2M 5.6M 7M SE +/- 92751.45, N = 3 SE +/- 32592.10, N = 3 SE +/- 63949.45, N = 3 SE +/- 6192.01, N = 12 6607012 6492204 6293617 437650 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 500 1000 1500 2000 2500 SE +/- 11.40, N = 3 SE +/- 9.03, N = 3 SE +/- 8.20, N = 3 SE +/- 6.78, N = 4 2539 2533 2324 427
Rodinia Rodinia is a suite focused upon accelerating compute-intensive applications with accelerators. CUDA, OpenMP, and OpenCL parallel models are supported by the included applications. This profile utilizes the OpenCL and OpenMP test binaries at the moment. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 20 40 60 80 100 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 1.05, N = 3 28.57 29.60 29.77 109.13 1. (CXX) g++ options: -O2 -lOpenCL
PHPBench PHPBench is a benchmark suite for PHP. It performs a large number of simple tests in order to bench various aspects of the PHP interpreter. PHPBench can be used to compare hardware, operating systems, PHP versions, PHP accelerators and caches, compiler options, etc. The number of iterations used is 1,000,000. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite Kernel Protection No Protection OldFart Fedora Default Kernel + User Protection 70K 140K 210K 280K 350K SE +/- 554.58, N = 3 SE +/- 3520.86, N = 3 SE +/- 606.42, N = 3 SE +/- 1702.94, N = 3 330044 329783 168448 159731
Bork File Encrypter Bork is a small, cross-platform file encryption utility. It is written in Java and designed to be included along with the files it encrypts for long-term storage. This test measures the amount of time it takes to encrypt a sample file. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Seconds, Fewer Is Better Bork File Encrypter 1.4 File Encryption Time No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.04, N = 3 SE +/- 0.02, N = 3 SE +/- 0.42, N = 9 22.16 22.22 22.51 36.26
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: http OldFart Fedora Default Kernel + User Protection No Protection Kernel Protection 400K 800K 1200K 1600K 2000K SE +/- 183.50, N = 3 SE +/- 107637.64, N = 12 SE +/- 209033.95, N = 9 SE +/- 220954.01, N = 10 54528 1341717 1568381 1928510
Stress-NG Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Semaphores OldFart Fedora No Protection Default Kernel + User Protection Kernel Protection 400K 800K 1200K 1600K 2000K SE +/- 25921.47, N = 4 SE +/- 8664.98, N = 3 SE +/- 4183.35, N = 3 SE +/- 5258.60, N = 3 1694586 456638 347517 347166 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Memory Copying No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 2K 4K 6K 8K 10K SE +/- 154.42, N = 3 SE +/- 154.58, N = 3 SE +/- 122.58, N = 3 SE +/- 0.25, N = 3 8368 8340 8317 203 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: CPU Stress No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 3K 6K 9K 12K 15K SE +/- 9.31, N = 3 SE +/- 18.47, N = 3 SE +/- 7.41, N = 3 SE +/- 0.85, N = 3 12402 12321 12279 284 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Tsearch No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 200 400 600 800 1000 SE +/- 2.73, N = 3 SE +/- 2.77, N = 3 SE +/- 5.85, N = 3 SE +/- 0.01, N = 3 898.00 889.00 859.00 30.95 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Forking No Protection Default Kernel + User Protection Kernel Protection OldFart Fedora 13K 26K 39K 52K 65K SE +/- 278.00, N = 3 SE +/- 106.66, N = 3 SE +/- 643.64, N = 3 SE +/- 36.41, N = 3 60585 59360 58592 4864 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Socket Activity No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 6K 12K 18K 24K 30K SE +/- 211.09, N = 3 SE +/- 103.85, N = 3 SE +/- 143.39, N = 3 SE +/- 5.27, N = 3 28717 27094 26272 308 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Crypto No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 2K 4K 6K 8K 10K SE +/- 11.73, N = 3 SE +/- 66.57, N = 3 SE +/- 41.59, N = 3 SE +/- 1.28, N = 3 8298 8149 8075 237 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Vector Math No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 20K 40K 60K 80K 100K SE +/- 225.56, N = 3 SE +/- 212.84, N = 3 SE +/- 114.43, N = 3 SE +/- 0.80, N = 3 88699 88416 88154 1747 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Glibc C String Functions No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 300K 600K 900K 1200K 1500K SE +/- 14521.20, N = 3 SE +/- 8505.49, N = 3 SE +/- 6341.64, N = 3 SE +/- 264.34, N = 3 1379226 1370824 1290587 31256 -laio -laio -laio 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -lc
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: garbage No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 3M 6M 9M 12M 15M SE +/- 3498.26, N = 3 SE +/- 12454.68, N = 12 SE +/- 12597.46, N = 12 SE +/- 51413.76, N = 3 1106825 1211245 1233700 12861996
OpenSSL OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance Kernel Protection Default Kernel + User Protection No Protection OldFart Fedora 1600 3200 4800 6400 8000 SE +/- 23.66, N = 3 SE +/- 51.34, N = 3 SE +/- 26.62, N = 3 SE +/- 0.32, N = 3 7417 7394 7390 200 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Go Benchmarks Benchmark for monitoring real time performance of the Go implementation for HTTP, JSON and garbage testing per iteration. Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org Nanoseconds/Operation, Fewer Is Better Go Benchmarks Test: json No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 12M 24M 36M 48M 60M SE +/- 54441.22, N = 3 SE +/- 3340.75, N = 3 SE +/- 87285.98, N = 12 SE +/- 947223.24, N = 3 3813471 4238819 4898644 57034155
FFTE FFTE is a package by Daisuke Takahashi to compute Discrete Fourier Transforms of 1-, 2- and 3- dimensional sequences of length (2^p)*(3^q)*(5^r). Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MFLOPS, More Is Better FFTE 6.0 Test: N=256, 1D Complex FFT Routine Kernel Protection 1500 3000 4500 6000 7500 SE +/- 0.43, N = 3 6770 1. (F9X) gfortran options: -O3 -fomit-frame-pointer -fopenmp -pthread -lmpi_usempif08 -lmpi_mpifh -lmpi
Compile Bench Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Read Compiled Tree No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 400 800 1200 1600 2000 SE +/- 0.59, N = 3 SE +/- 7.11, N = 3 SE +/- 4.49, N = 3 SE +/- 1.19, N = 3 1887 1823 1538 315
OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create No Protection Kernel Protection Default Kernel + User Protection OldFart Fedora 80 160 240 320 400 SE +/- 1.84, N = 3 SE +/- 0.18, N = 3 SE +/- 1.65, N = 3 SE +/- 1.53, N = 3 348.00 346.00 296.00 96.59
Kernel Protection Compiler Notes: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -vProcessor Notes: Scaling Governor: powernv-cpufreq ondemandPython Notes: Python 2.7.15+ + Python 3.6.7Security Notes: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch serialisation (kernel only) + Kernel entry/exit barrier (eieio)
Testing initiated at 8 November 2018 09:49 by user phoronix.
Default Kernel + User Protection Compiler Notes: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: powernv-cpufreq ondemandPython Notes: Python 2.7.15+ + Python 3.6.7Security Notes: __user pointer sanitization ori31 speculation barrier enabled + Indirect branch cache disabled + Kernel entry/exit barrier (eieio)
Testing initiated at 8 November 2018 19:48 by user phoronix.
No Protection Processor: POWER9 altivec supported @ 3.80GHz (44 Cores / 176 Threads), Motherboard: PowerNV T2P9D01 REV 1.01, Memory: 65536MB, Disk: Samsung SSD 960 EVO 500GB, Graphics: ASPEED ASPEED Family, Monitor: VE228, Network: Broadcom NetXtreme BCM5719 Gigabit PCIe
OS: Ubuntu 18.10, Kernel: 4.18.0-10-generic (ppc64le), Compiler: GCC 8.2.0, File-System: ext4, Screen Resolution: 1920x1080
Compiler Notes: --build=powerpc64le-linux-gnu --disable-libphobos --disable-multilib --disable-werror --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++ --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-nls --enable-objc-gc=auto --enable-plugin --enable-secureplt --enable-shared --enable-targets=powerpcle-linux --enable-threads=posix --host=powerpc64le-linux-gnu --program-prefix=powerpc64le-linux-gnu- --target=powerpc64le-linux-gnu --with-cpu=power8 --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-long-double-128 -vDisk Notes: NONE / errors=remount-ro,relatime,rwProcessor Notes: Scaling Governor: powernv-cpufreq ondemandPython Notes: Python 2.7.15+ + Python 3.6.7
Testing initiated at 9 November 2018 04:15 by user phoronix.
OldFart Fedora Processor: AMD A6-3410MX APU @ 1.60GHz (4 Cores), Motherboard: HP 358D v33.18 (F.21 BIOS), Chipset: AMD Family 12h Root Complex, Memory: 6144MB, Disk: 1000GB Samsung SSD 840, Graphics: AMD Radeon HD 6520G 512MB, Audio: AMD BeaverCreek HDMI Audio, Network: Realtek RTL8111/8168/8411 + Broadcom and subsidiaries BCM4313 802.11bgn Wireless
OS: Fedora 29, Kernel: 4.19.1jan (x86_64), Desktop: GNOME Shell 3.30.1, Display Server: Wayland, OpenGL: 3.3 Mesa 18.2.4 (LLVM 7.0.0), File-System: ext4, Screen Resolution: 1366x768
Disk Notes: BFQ / relatime,rw,seclabelProcessor Notes: Scaling Governor: acpi-cpufreq ondemandPython Notes: Python 2.7.15 + Python 3.7.1Security Notes: SELinux + __user pointer sanitization + Full AMD retpoline
Testing initiated at 13 November 2018 14:17 by user jan.