Gentoo Kernel Tuning Patch Benchmarks AMD Ryzen 7 2700X Eight-Core testing with a ASUS ROG CROSSHAIR VII HERO (WI-FI) (1001 BIOS) and Sapphire AMD Radeon RX 470/480/570/570X/580/580X 4GB on Ubuntu 18.10 via the Phoronix Test Suite.
HTML result view exported from: https://openbenchmarking.org/result/1902251-PTS-GENTOOPA35&grs&sor .
Processor Motherboard Chipset Memory Disk Graphics Audio Monitor Network OS Kernel Desktop Display Server Display Driver OpenGL Compiler File-System Screen Resolution i9-7980XE 2700X Stock Tuned Stock Tuned Intel Core i9-7980XE @ 4.20GHz (18 Cores / 36 Threads) ASUS PRIME X299-A (1602 BIOS) Intel Sky Lake-E DMI3 Registers 16384MB Samsung SSD 970 EVO 500GB NVIDIA NV120 12GB Realtek ALC1220 ASUS PB278 Intel I219-V Ubuntu 18.10 5.0.0-rc7-stock (x86_64) 20190224 GNOME Shell 3.30.1 X Server 1.20.1 modesetting 1.20.1 4.3 Mesa 18.2.2 GCC 8.2.0 ext4 2560x1440 5.0.0-rc7-skylakex (x86_64) 20190224 AMD Ryzen 7 2700X Eight-Core @ 3.70GHz (8 Cores / 16 Threads) ASUS ROG CROSSHAIR VII HERO (WI-FI) (1001 BIOS) AMD Family 17h Samsung SSD 970 EVO 250GB Sapphire AMD Radeon RX 470/480/570/570X/580/580X 4GB (1260/1750MHz) AMD Ellesmere Intel I211 + Realtek RTL8822BE 802.11a/b/g/n/ac 5.0.0-rc7-stock (x86_64) 20190224 4.5 Mesa 18.2.2 (LLVM 7.0.0) 5.0.0-rc7-znver1 (x86_64) 20190224 OpenBenchmarking.org Compiler 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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib --with-tune=generic --without-cuda-driver -v Disk Details - NONE / errors=remount-ro,relatime,rw Processor Details - i9-7980XE: Stock: Scaling Governor: intel_pstate powersave - i9-7980XE: Tuned: Scaling Governor: intel_pstate powersave - 2700X: Stock: Scaling Governor: acpi-cpufreq ondemand - 2700X: Tuned: Scaling Governor: acpi-cpufreq ondemand Python Details - Python 2.7.15+ + Python 3.6.7 Security Details - i9-7980XE: Stock: KPTI + __user pointer sanitization + Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + SSB disabled via prctl and seccomp + PTE Inversion; VMX: conditional cache flushes SMT vulnerable - i9-7980XE: Tuned: KPTI + __user pointer sanitization + Full generic retpoline IBPB: conditional IBRS_FW STIBP: conditional RSB filling + SSB disabled via prctl and seccomp + PTE Inversion; VMX: conditional cache flushes SMT vulnerable - 2700X: Stock: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp - 2700X: Tuned: __user pointer sanitization + Full AMD retpoline IBPB: conditional STIBP: disabled RSB filling + SSB disabled via prctl and seccomp
svt-vp9: 1080p 8-bit YUV To VP9 Video Encode openssl: RSA 4096-bit Performance compress-zstd: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 pgbench: Buffer Test - Normal Load - Read Only namd: ATPase Simulation - 327,506 Atoms compress-7zip: Compress Speed Test rodinia: OpenMP CFD Solver himeno: Poisson Pressure Solver build-llvm: Time To Compile tinymembench: Standard Memset cloverleaf: Lagrangian-Eulerian Hydrodynamics brl-cad: VGR Performance Metric svt-hevc: 1080p 8-bit YUV To HEVC Video Encode ctx-clock: Context Switch Time dacapobench: Tradebeans blender: BMW27 - CPU-Only ttsiod-renderer: Phong Rendering With Soft-Shadow Mapping stress-ng: Forking build-linux-kernel: Time To Compile parboil: OpenMP LBM rodinia: OpenMP LavaMD sockperf: Throughput hackbench: 32 - Process apache-siege: 100 stress-ng: Vector Math dacapobench: H2 hackbench: 16 - Process tinymembench: Standard Memcpy stress-ng: Crypto c-ray: Total Time - 4K, 16 Rays Per Pixel mbw: Memory Copy - 4096 MiB stress-ng: System V Message Passing stress-ng: Semaphores dacapobench: Jython parboil: OpenMP CUTCP rodinia: OpenMP Streamcluster cryptsetup: PBKDF2-sha512 compilebench: Initial Create phpbench: PHP Benchmark Suite apache-siege: 50 dacapobench: Tradesoap octave-benchmark: fs-mark: 1000 Files, 1MB Size cryptsetup: PBKDF2-whirlpool pybench: Total For Average Test Times node-express-loadtest: apache-siege: 10 vpxenc: vpxenc VP9 1080p Video Encode cryptsetup: PBKDF2-whirlpool fio: Seq Read - Linux AIO - Yes - No - 2MB - Default Test Directory node-octane: fio: Rand Read - Linux AIO - Yes - No - 2MB - Default Test Directory tesseract-ocr: Time To OCR 7 Images sqlite: Timed SQLite Insertions opencv-bench: luajit: Composite pgbench: Buffer Test - Normal Load - Read Write scikit-learn: mbw: Memory Copy, Fixed Block Size - 4096 MiB compress-xz: Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 graphics-magick: Resizing xonotic: 2560 x 1440 - Ultra nginx: Static Web Page Serving radiance: SMP Parallel xonotic: 2560 x 1440 - Ultimate fio: Rand Write - Linux AIO - Yes - No - 2MB - Default Test Directory git: Time To Complete Common Git Commands fio: Seq Write - Linux AIO - Yes - No - 2MB - Default Test Directory etlegacy: Renderer2 - 2560 x 1440 openarena: 2560 x 1440 paraview: Wavelet Volume - 2560 x 1440 tesseract: 2560 x 1440 rbenchmark: paraview: Wavelet Volume - 2560 x 1440 paraview: Many Spheres - 2560 x 1440 paraview: Wavelet Contour - 2560 x 1440 paraview: Many Spheres - 2560 x 1440 cython-bench: glmark2: 2560 x 1440 paraview: Wavelet Contour - 2560 x 1440 sunflow: Global Illumination + Image Synthesis apache-siege: 250 apache-siege: 200 stress-ng: Context Switching stress-ng: Socket Activity hackbench: 8 - Process hackbench: 4 - Process hackbench: 16 - Thread hackbench: 8 - Thread hackbench: 4 - Thread sockperf: Latency Under Load sockperf: Latency Ping Pong compilebench: Compile fs-mark: 4000 Files, 32 Sub Dirs, 1MB Size i9-7980XE 2700X Stock Tuned Stock Tuned 309 4689 10.44 472297 0.98351 94645 11.58 3102 233 28545 2.43 210187 309 372 8812 116 708 105603 45.70 73.01 20.54 303563 43.36 43448 41008 6068 22.45 10298 3575 42.42 10539 7890690 3023293 3591 2.38 13.87 1661085 665 705214 35305 4826 17.68 189 983654 948 9536 25338 22.84 7689 45426 7670 26.49 39.57 79.84 1622 5449 13.60 9969 84.96 247 35324 239 4782 5.41 4726 0.2928 20.43 1.01 46175 38182 20723337 15571 12.23 6.64 36.36 20.11 9.08 36.56 4.60 1466 113 315 4690 10.44 469329 0.98252 94958 11.56 3093 234 28698 2.40 210634 310 372 8741 116 709 105655 45.65 72.90 20.25 304632 43.27 43840 41002 5973 23.56 10358 3578 42.39 10545 7743369 3228754 3661 2.38 14.18 1671601 686 705433 35384 4807 17.65 190 983654 949 9590 24166 22.80 981223 7445 45197 7783 26.37 39.73 79.70 1627 5470 13.60 10057 85.11 247 35619 237 4847 5.40 4749 0.2915 20.45 0.98 40011 36038 20720973 15637 13.58 6.65 35.16 20.79 8.82 39.68 4.40 1525 113 52.93 1702 28.37 189687 2.29339 40730 26.89 1334 527 12789 5.06 101629 150 185 4392 230 363 54253 86.52 138.03 37.56 510316 77.14 32145 23616 3530 36.86 16689 2208 66.11 16171 11628479 4409946 5169 3.23 18.90 1256293 525 539407 29026 5860 14.69 226 823169 1129 8336 26972 26.60 846080 8419 40771 8442 28.86 36.33 74.27 1537 5711 14.30 9584 88.86 256 325 34671 243 258 4860 5.32 4796 211 271 149 341 2391 3306 1387 32.98 2730 133 1.46 34025 24403 3880010 7399 18.13 11.45 38.75 25.45 14.39 106.96 4.20 916 131 52.86 1705 28.37 191178 2.29314 41000 26.78 1357 534 12781 5.03 100880 150 185 4409 230 364 54104 86.91 137.98 37.47 559386 77.05 24877 23718 3665 36.23 16704 2209 66.08 16177 11581693 4389305 5038 3.28 18.86 1234877 522 541769 27389 5836 14.63 224 846308 1131 8180 28304 26.54 844719 8459 40723 8389 28.77 36.40 73.86 1518 5350 14.18 9577 88.74 257 316 34957 241 252 4876 5.32 4792 209 269 150 339 2397 3314 1384 33.05 2732 133 1.45 34231 31160 3700685 7715 16.48 10.96 37.53 23.57 13.96 106.65 4.25 965 103 OpenBenchmarking.org
SVT-VP9 1080p 8-bit YUV To VP9 Video Encode i9-7980XE 2700X OpenBenchmarking.org Frames Per Second, More Is Better SVT-VP9 2019-02-17 1080p 8-bit YUV To VP9 Video Encode Tuned Stock 70 140 210 280 350 SE +/- 5.69, N = 3 SE +/- 4.75, N = 3 SE +/- 0.08, N = 3 SE +/- 0.05, N = 3 315.00 309.00 52.93 52.86 1. (CC) gcc options: -fPIE -fPIC -O2 -flto -fvisibility=hidden -mavx -pie -rdynamic -lpthread -lrt -lm
OpenArena Resolution: 2560 x 1440 - Total Frame Time OpenBenchmarking.org Milliseconds, Fewer Is Better OpenArena 0.8.8 Resolution: 2560 x 1440 - Total Frame Time Stock Tuned 6 12 18 24 30 Min: 1 / Avg: 3.66 / Max: 26 Min: 1 / Avg: 3.7 / Max: 25
OpenSSL RSA 4096-bit Performance i9-7980XE 2700X OpenBenchmarking.org Signs Per Second, More Is Better OpenSSL 1.1.1 RSA 4096-bit Performance Tuned Stock 1000 2000 3000 4000 5000 SE +/- 4.65, N = 3 SE +/- 2.78, N = 3 SE +/- 4.09, N = 3 SE +/- 2.34, N = 3 4690 4689 1705 1702 1. (CC) gcc options: -pthread -m64 -O3 -lssl -lcrypto -ldl
Zstd Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Zstd Compression 1.3.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 19 Stock Tuned 7 14 21 28 35 SE +/- 0.02, N = 3 SE +/- 0.05, N = 3 SE +/- 0.07, N = 3 SE +/- 0.03, N = 3 10.44 10.44 28.37 28.37 1. (CC) gcc options: -O3 -pthread -lz
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Only i9-7980XE 2700X OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Only Stock Tuned 100K 200K 300K 400K 500K SE +/- 1128.42, N = 3 SE +/- 651.72, N = 3 SE +/- 112.88, N = 3 SE +/- 674.60, N = 3 472297 469329 191178 189687 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
NAMD ATPase Simulation - 327,506 Atoms i9-7980XE 2700X OpenBenchmarking.org days/ns, Fewer Is Better NAMD 2.13b1 ATPase Simulation - 327,506 Atoms Tuned Stock 0.516 1.032 1.548 2.064 2.58 SE +/- 0.00063, N = 3 SE +/- 0.00328, N = 3 SE +/- 0.00103, N = 3 SE +/- 0.00091, N = 3 0.98252 0.98351 2.29314 2.29339
7-Zip Compression Compress Speed Test i9-7980XE 2700X OpenBenchmarking.org MIPS, More Is Better 7-Zip Compression 16.02 Compress Speed Test Tuned Stock 20K 40K 60K 80K 100K SE +/- 276.27, N = 3 SE +/- 337.14, N = 3 SE +/- 77.18, N = 3 SE +/- 483.13, N = 3 94958 94645 41000 40730 1. (CXX) g++ options: -pipe -lpthread
Rodinia Test: OpenMP CFD Solver i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP CFD Solver Tuned Stock 6 12 18 24 30 SE +/- 0.03, N = 3 SE +/- 0.02, N = 3 SE +/- 0.06, N = 3 SE +/- 0.03, N = 3 11.56 11.58 26.78 26.89 1. (CXX) g++ options: -O2 -lOpenCL
Himeno Benchmark Poisson Pressure Solver i9-7980XE 2700X OpenBenchmarking.org MFLOPS, More Is Better Himeno Benchmark 3.0 Poisson Pressure Solver Stock Tuned 700 1400 2100 2800 3500 SE +/- 5.10, N = 3 SE +/- 1.43, N = 3 SE +/- 1.83, N = 3 SE +/- 7.16, N = 3 3102 3093 1357 1334 1. (CC) gcc options: -O3 -mavx2
Timed LLVM Compilation Time To Compile i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Timed LLVM Compilation 6.0.1 Time To Compile Stock Tuned 120 240 360 480 600 233 234 527 534
Tinymembench Standard Memset i9-7980XE 2700X OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memset Tuned Stock 6K 12K 18K 24K 30K SE +/- 16.11, N = 3 SE +/- 142.56, N = 3 SE +/- 17.50, N = 3 SE +/- 9.34, N = 3 28698 28545 12789 12781 1. (CC) gcc options: -O2 -lm
CloverLeaf Lagrangian-Eulerian Hydrodynamics i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better CloverLeaf Lagrangian-Eulerian Hydrodynamics Tuned Stock 1.1385 2.277 3.4155 4.554 5.6925 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.00, N = 3 SE +/- 0.00, N = 3 2.40 2.43 5.03 5.06 1. (F9X) gfortran options: -O3 -march=native -funroll-loops -fopenmp
BRL-CAD VGR Performance Metric i9-7980XE 2700X OpenBenchmarking.org VGR Performance Metric, More Is Better BRL-CAD 7.28.0 VGR Performance Metric Tuned Stock 50K 100K 150K 200K 250K 210634 210187 101629 100880 1. (CXX) g++ options: -std=c++98 -pipe -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -rdynamic -lpthread -ldl -lm
SVT-HEVC 1080p 8-bit YUV To HEVC Video Encode i9-7980XE 2700X OpenBenchmarking.org Frames Per Second, More Is Better SVT-HEVC 2019-02-03 1080p 8-bit YUV To HEVC Video Encode Tuned Stock 70 140 210 280 350 SE +/- 4.13, N = 3 SE +/- 3.85, N = 3 SE +/- 0.71, N = 3 SE +/- 0.53, N = 3 310 309 150 150 1. (CC) gcc options: -fPIE -fPIC -O2 -flto -fvisibility=hidden -march=native -pie -rdynamic -lpthread -lrt
ctx_clock Context Switch Time 2700X i9-7980XE OpenBenchmarking.org Clocks, Fewer Is Better ctx_clock Context Switch Time Stock Tuned 80 160 240 320 400 185 185 372 372
DaCapo Benchmark Java Test: Tradebeans 2700X i9-7980XE OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradebeans Stock Tuned 2K 4K 6K 8K 10K SE +/- 19.38, N = 4 SE +/- 44.56, N = 4 SE +/- 24.72, N = 4 SE +/- 24.93, N = 4 4392 4409 8741 8812
Blender Blend File: BMW27 - Compute: CPU-Only i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Blender 2.79a Blend File: BMW27 - Compute: CPU-Only Stock Tuned 50 100 150 200 250 116 116 230 230
TTSIOD 3D Renderer Phong Rendering With Soft-Shadow Mapping i9-7980XE 2700X OpenBenchmarking.org FPS, More Is Better TTSIOD 3D Renderer 2.3b Phong Rendering With Soft-Shadow Mapping Tuned Stock 150 300 450 600 750 SE +/- 1.36, N = 3 SE +/- 2.52, N = 3 SE +/- 0.57, N = 3 SE +/- 0.65, N = 3 709 708 364 363 1. (CXX) g++ options: -O3 -fomit-frame-pointer -ffast-math -mtune=native -flto -msse -mrecip -mfpmath=sse -msse2 -mssse3 -lSDL -fopenmp -fwhole-program -lstdc++
Stress-NG Test: Forking i9-7980XE 2700X OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Forking Tuned Stock 20K 40K 60K 80K 100K SE +/- 341.77, N = 3 SE +/- 307.30, N = 3 SE +/- 119.31, N = 3 SE +/- 206.52, N = 3 105655 105603 54253 54104 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Timed Linux Kernel Compilation Time To Compile i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Timed Linux Kernel Compilation 4.18 Time To Compile Tuned Stock 20 40 60 80 100 SE +/- 0.62, N = 6 SE +/- 0.60, N = 6 SE +/- 0.95, N = 3 SE +/- 0.75, N = 3 45.65 45.70 86.52 86.91
Parboil Test: OpenMP LBM i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP LBM Tuned Stock 30 60 90 120 150 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 72.90 73.01 137.98 138.03 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Rodinia Test: OpenMP LavaMD i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP LavaMD Tuned Stock 9 18 27 36 45 SE +/- 0.04, N = 3 SE +/- 0.28, N = 3 SE +/- 0.07, N = 3 SE +/- 0.05, N = 3 20.25 20.54 37.47 37.56 1. (CXX) g++ options: -O2 -lOpenCL
Sockperf Test: Throughput 2700X i9-7980XE OpenBenchmarking.org Messages Per Second, More Is Better Sockperf 3.4 Test: Throughput Tuned Stock 120K 240K 360K 480K 600K SE +/- 6925.66, N = 5 SE +/- 6168.72, N = 5 SE +/- 3358.81, N = 5 SE +/- 3597.48, N = 5 559386 510316 304632 303563 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Hackbench Count: 32 - Type: Process i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 32 - Type: Process Tuned Stock 20 40 60 80 100 SE +/- 0.11, N = 3 SE +/- 0.68, N = 3 SE +/- 1.27, N = 3 SE +/- 1.51, N = 3 43.27 43.36 77.05 77.14 1. (CC) gcc options: -lpthread
Apache Siege Concurrent Users: 100 i9-7980XE 2700X OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 100 Tuned Stock 9K 18K 27K 36K 45K SE +/- 613.13, N = 6 SE +/- 438.59, N = 11 SE +/- 371.37, N = 12 SE +/- 299.99, N = 3 43840 43448 32145 24877 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Stress-NG Test: Vector Math i9-7980XE 2700X OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Vector Math Stock Tuned 9K 18K 27K 36K 45K SE +/- 6.64, N = 3 SE +/- 2.17, N = 3 SE +/- 0.37, N = 3 SE +/- 11.56, N = 3 41008 41002 23718 23616 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
DaCapo Benchmark Java Test: H2 2700X i9-7980XE OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: H2 Stock Tuned 1300 2600 3900 5200 6500 SE +/- 40.54, N = 9 SE +/- 36.79, N = 4 SE +/- 47.06, N = 4 SE +/- 28.41, N = 4 3530 3665 5973 6068
Hackbench Count: 16 - Type: Process i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Process Stock Tuned 8 16 24 32 40 SE +/- 0.31, N = 12 SE +/- 0.31, N = 12 SE +/- 0.60, N = 12 SE +/- 0.25, N = 3 22.45 23.56 36.23 36.86 1. (CC) gcc options: -lpthread
Tinymembench Standard Memcpy 2700X i9-7980XE OpenBenchmarking.org MB/s, More Is Better Tinymembench 2018-05-28 Standard Memcpy Tuned Stock 4K 8K 12K 16K 20K SE +/- 9.74, N = 3 SE +/- 17.81, N = 3 SE +/- 2.38, N = 3 SE +/- 40.77, N = 3 16704 16689 10358 10298 1. (CC) gcc options: -O2 -lm
Stress-NG Test: Crypto i9-7980XE 2700X OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Crypto Tuned Stock 800 1600 2400 3200 4000 SE +/- 1.53, N = 3 SE +/- 0.40, N = 3 SE +/- 1.28, N = 3 SE +/- 1.65, N = 3 3578 3575 2209 2208 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
C-Ray Total Time - 4K, 16 Rays Per Pixel i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better C-Ray 1.1 Total Time - 4K, 16 Rays Per Pixel Tuned Stock 15 30 45 60 75 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.08, N = 3 SE +/- 0.03, N = 3 42.39 42.42 66.08 66.11 1. (CC) gcc options: -lm -lpthread -O3
MBW Test: Memory Copy - Array Size: 4096 MiB 2700X i9-7980XE OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 4096 MiB Tuned Stock 3K 6K 9K 12K 15K SE +/- 6.89, N = 3 SE +/- 14.43, N = 3 SE +/- 30.47, N = 3 SE +/- 0.73, N = 3 16177 16171 10545 10539 1. (CC) gcc options: -O3 -march=native
Stress-NG Test: System V Message Passing 2700X i9-7980XE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: System V Message Passing Stock Tuned 2M 4M 6M 8M 10M SE +/- 30665.10, N = 3 SE +/- 25682.03, N = 3 SE +/- 2591.95, N = 3 SE +/- 4195.33, N = 3 11628479 11581693 7890690 7743369 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Semaphores 2700X i9-7980XE OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Semaphores Stock Tuned 900K 1800K 2700K 3600K 4500K SE +/- 38215.81, N = 3 SE +/- 10286.68, N = 3 SE +/- 3336.75, N = 3 SE +/- 8724.64, N = 3 4409946 4389305 3228754 3023293 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
DaCapo Benchmark Java Test: Jython i9-7980XE 2700X OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Jython Stock Tuned 1100 2200 3300 4400 5500 SE +/- 12.64, N = 4 SE +/- 31.69, N = 4 SE +/- 31.46, N = 4 3591 3661 5038 5169
Parboil Test: OpenMP CUTCP i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Parboil 2.5 Test: OpenMP CUTCP Stock Tuned 0.738 1.476 2.214 2.952 3.69 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.02, N = 3 SE +/- 0.01, N = 3 2.38 2.38 3.23 3.28 1. (CXX) g++ options: -lm -lpthread -lgomp -O3 -ffast-math -fopenmp
Rodinia Test: OpenMP Streamcluster i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Rodinia 2.4 Test: OpenMP Streamcluster Stock Tuned 5 10 15 20 25 SE +/- 0.02, N = 3 SE +/- 0.20, N = 6 SE +/- 0.00, N = 3 SE +/- 0.01, N = 3 13.87 14.18 18.86 18.90 1. (CXX) g++ options: -O2 -lOpenCL
Cryptsetup PBKDF2-sha512 i9-7980XE 2700X OpenBenchmarking.org Iterations Per Second, More Is Better Cryptsetup 2.0.4 PBKDF2-sha512 Tuned Stock 400K 800K 1200K 1600K 2000K SE +/- 890.33, N = 3 SE +/- 10002.19, N = 3 SE +/- 12644.36, N = 3 SE +/- 878.73, N = 3 SE +/- 1337.36, N = 3 SE +/- 2799.15, N = 3 SE +/- 18903.89, N = 4 1673260 1671601 1661085 1263345 1261321 1256293 1234877
Compile Bench Test: Initial Create i9-7980XE 2700X OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Initial Create Tuned Stock 150 300 450 600 750 SE +/- 9.75, N = 3 SE +/- 15.33, N = 3 SE +/- 1.22, N = 3 SE +/- 4.29, N = 3 686 665 525 522
PHPBench PHP Benchmark Suite i9-7980XE 2700X OpenBenchmarking.org Score, More Is Better PHPBench 0.8.1 PHP Benchmark Suite Tuned Stock 150K 300K 450K 600K 750K SE +/- 532.80, N = 3 SE +/- 1083.13, N = 3 SE +/- 2174.06, N = 3 SE +/- 953.97, N = 3 705433 705214 541769 539407
Apache Siege Concurrent Users: 50 i9-7980XE 2700X OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 50 Tuned Stock 8K 16K 24K 32K 40K SE +/- 513.39, N = 3 SE +/- 579.95, N = 3 SE +/- 574.67, N = 3 SE +/- 301.55, N = 3 35384 35305 29026 27389 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
DaCapo Benchmark Java Test: Tradesoap i9-7980XE 2700X OpenBenchmarking.org msec, Fewer Is Better DaCapo Benchmark 9.12-MR1 Java Test: Tradesoap Tuned Stock 1300 2600 3900 5200 6500 SE +/- 12.98, N = 4 SE +/- 33.71, N = 4 SE +/- 58.63, N = 4 SE +/- 54.36, N = 4 4807 4826 5836 5860
GNU Octave Benchmark 2700X i9-7980XE OpenBenchmarking.org Seconds, Fewer Is Better GNU Octave Benchmark 4.4.1 Tuned Stock 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 SE +/- 0.06, N = 3 SE +/- 0.04, N = 3 14.63 14.69 17.65 17.68
FS-Mark Test: 1000 Files, 1MB Size 2700X i9-7980XE OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 1000 Files, 1MB Size Stock Tuned 50 100 150 200 250 SE +/- 0.38, N = 3 SE +/- 0.15, N = 3 SE +/- 2.88, N = 3 SE +/- 2.59, N = 3 226 224 190 189 1. (CC) gcc options: -static
Cryptsetup PBKDF2-whirlpool i9-7980XE 2700X OpenBenchmarking.org Iterations Per Second, More Is Better Cryptsetup PBKDF2-whirlpool Tuned Stock 200K 400K 600K 800K 1000K SE +/- 394.33, N = 4 SE +/- 19755.34, N = 3 983654 983654 846308 823169
PyBench Total For Average Test Times i9-7980XE 2700X OpenBenchmarking.org Milliseconds, Fewer Is Better PyBench 2018-02-16 Total For Average Test Times Stock Tuned 200 400 600 800 1000 SE +/- 1.53, N = 3 SE +/- 1.76, N = 3 SE +/- 2.33, N = 3 SE +/- 4.00, N = 3 948 949 1129 1131
Node.js Express HTTP Load Test i9-7980XE 2700X OpenBenchmarking.org Requests Per Second, More Is Better Node.js Express HTTP Load Test Tuned Stock 2K 4K 6K 8K 10K SE +/- 25.12, N = 3 SE +/- 6.81, N = 3 SE +/- 68.75, N = 3 SE +/- 118.06, N = 5 9590 9536 8336 8180 1. Nodejs
v8.11.4
Apache Siege Concurrent Users: 10 2700X i9-7980XE OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 10 Tuned Stock 6K 12K 18K 24K 30K SE +/- 185.69, N = 3 SE +/- 341.78, N = 12 SE +/- 56.69, N = 3 SE +/- 368.29, N = 3 28304 26972 25338 24166 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
VP9 libvpx Encoding vpxenc VP9 1080p Video Encode 2700X i9-7980XE OpenBenchmarking.org Frames Per Second, More Is Better VP9 libvpx Encoding 1.8.0 vpxenc VP9 1080p Video Encode Stock Tuned 6 12 18 24 30 SE +/- 0.25, N = 3 SE +/- 0.38, N = 3 SE +/- 0.04, N = 3 SE +/- 0.07, N = 3 26.60 26.54 22.84 22.80 1. (CXX) g++ options: -m64 -lm -lpthread -O3 -fPIC -U_FORTIFY_SOURCE -std=c++11
Cryptsetup PBKDF2-whirlpool i9-7980XE 2700X OpenBenchmarking.org Iterations Per Second, More Is Better Cryptsetup 2.0.4 PBKDF2-whirlpool Tuned Stock 200K 400K 600K 800K 1000K SE +/- 3397.36, N = 3 SE +/- 455.33, N = 3 SE +/- 906.33, N = 3 981223 846080 844719
Flexible IO Tester Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory 2700X i9-7980XE OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory Tuned Stock 2K 4K 6K 8K 10K SE +/- 32.64, N = 3 SE +/- 14.26, N = 3 SE +/- 22.01, N = 3 SE +/- 95.75, N = 12 8459 8419 7689 7445 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -ll -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Node.js Octane Benchmark i9-7980XE 2700X OpenBenchmarking.org Score, More Is Better Node.js Octane Benchmark Stock Tuned 10K 20K 30K 40K 50K SE +/- 43.68, N = 3 SE +/- 128.44, N = 3 SE +/- 50.95, N = 3 SE +/- 138.17, N = 3 45426 45197 40771 40723 1. Nodejs
v8.11.4
Flexible IO Tester Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory 2700X i9-7980XE OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Read - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory Stock Tuned 2K 4K 6K 8K 10K SE +/- 58.44, N = 3 SE +/- 39.73, N = 3 SE +/- 29.26, N = 3 SE +/- 22.03, N = 3 8442 8389 7783 7670 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -ll -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Tesseract OCR Time To OCR 7 Images i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Tesseract OCR 4.0.0-beta.3-249-g607e Time To OCR 7 Images Tuned Stock 7 14 21 28 35 SE +/- 0.09, N = 3 SE +/- 0.04, N = 3 SE +/- 0.05, N = 3 SE +/- 0.06, N = 3 26.37 26.49 28.77 28.86
SQLite Timed SQLite Insertions 2700X i9-7980XE OpenBenchmarking.org Seconds, Fewer Is Better SQLite 3.22 Timed SQLite Insertions Stock Tuned 9 18 27 36 45 SE +/- 0.01, N = 3 SE +/- 0.06, N = 3 SE +/- 0.01, N = 3 SE +/- 0.05, N = 3 36.33 36.40 39.57 39.73 1. (CC) gcc options: -O2 -lreadline -ltermcap -lz -ldl -lpthread
OpenCV Benchmark 2700X i9-7980XE OpenBenchmarking.org Seconds, Fewer Is Better OpenCV Benchmark 3.3.0 Tuned Stock 20 40 60 80 100 SE +/- 0.25, N = 3 SE +/- 0.15, N = 3 SE +/- 0.26, N = 3 SE +/- 0.39, N = 3 73.86 74.27 79.70 79.84 1. (CXX) g++ options: -std=c++11 -rdynamic
LuaJIT Test: Composite i9-7980XE 2700X OpenBenchmarking.org Mflops, More Is Better LuaJIT 2.1-git Test: Composite Tuned Stock 300 600 900 1200 1500 SE +/- 0.62, N = 3 SE +/- 0.74, N = 3 SE +/- 2.43, N = 3 SE +/- 2.21, N = 3 1627 1622 1537 1518 1. (CC) gcc options: -lm -ldl -O2 -fomit-frame-pointer -U_FORTIFY_SOURCE -fno-stack-protector
PostgreSQL pgbench Scaling: Buffer Test - Test: Normal Load - Mode: Read Write 2700X i9-7980XE OpenBenchmarking.org TPS, More Is Better PostgreSQL pgbench 10.3 Scaling: Buffer Test - Test: Normal Load - Mode: Read Write Stock Tuned 1200 2400 3600 4800 6000 SE +/- 83.19, N = 5 SE +/- 91.99, N = 3 SE +/- 39.82, N = 3 SE +/- 7.35, N = 3 5711 5350 5470 5449 1. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
Scikit-Learn i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Scikit-Learn 0.17.1 Stock Tuned 4 8 12 16 20 SE +/- 0.04, N = 3 SE +/- 0.04, N = 3 SE +/- 0.01, N = 3 SE +/- 0.13, N = 3 13.60 13.60 14.18 14.30
MBW Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB i9-7980XE 2700X OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiB Tuned Stock 2K 4K 6K 8K 10K SE +/- 0.75, N = 3 SE +/- 37.82, N = 3 SE +/- 3.78, N = 3 SE +/- 32.22, N = 3 10057 9969 9584 9577 1. (CC) gcc options: -O3 -march=native
XZ Compression Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better XZ Compression 5.2.4 Compressing ubuntu-16.04.3-server-i386.img, Compression Level 9 Stock Tuned 20 40 60 80 100 SE +/- 0.26, N = 3 SE +/- 0.20, N = 3 SE +/- 0.09, N = 3 SE +/- 0.14, N = 3 84.96 85.11 88.74 88.86 1. (CC) gcc options: -pthread -fvisibility=hidden -O2
GraphicsMagick Operation: Resizing 2700X i9-7980XE OpenBenchmarking.org Iterations Per Minute, More Is Better GraphicsMagick 1.3.30 Operation: Resizing Tuned Stock 60 120 180 240 300 257 256 247 247 1. (CC) gcc options: -fopenmp -O2 -pthread -ljbig -lwebp -lwebpmux -ltiff -ljpeg -lXext -lSM -lICE -lX11 -llzma -lbz2 -lz -lm -lgomp -lpthread
Xonotic Resolution: 2560 x 1440 - Effects Quality: Ultra OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 2560 x 1440 - Effects Quality: Ultra Stock Tuned 70 140 210 280 350 SE +/- 1.12, N = 3 SE +/- 3.69, N = 3 325 316 MIN: 184 / MAX: 632 MIN: 165 / MAX: 631
NGINX Benchmark Static Web Page Serving i9-7980XE 2700X OpenBenchmarking.org Requests Per Second, More Is Better NGINX Benchmark 1.9.9 Static Web Page Serving Tuned Stock 8K 16K 24K 32K 40K SE +/- 195.04, N = 3 SE +/- 115.86, N = 3 SE +/- 272.04, N = 3 SE +/- 197.44, N = 3 35619 35324 34957 34671 1. (CC) gcc options: -lpthread -lcrypt -lcrypto -lz -O3 -march=native
Radiance Benchmark Test: SMP Parallel i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Radiance Benchmark 5.0 Test: SMP Parallel Tuned Stock 50 100 150 200 250 237 239 241 243
Xonotic Resolution: 2560 x 1440 - Effects Quality: Ultimate OpenBenchmarking.org Frames Per Second, More Is Better Xonotic 0.8.2 Resolution: 2560 x 1440 - Effects Quality: Ultimate Stock Tuned 60 120 180 240 300 SE +/- 0.83, N = 3 SE +/- 0.98, N = 3 258 252 MIN: 47 / MAX: 504 MIN: 48 / MAX: 509
Flexible IO Tester Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory 2700X i9-7980XE OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Random Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory Tuned Stock 1000 2000 3000 4000 5000 SE +/- 16.95, N = 3 SE +/- 22.74, N = 3 SE +/- 24.51, N = 3 SE +/- 38.19, N = 3 4876 4860 4847 4782 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -ll -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
Git Time To Complete Common Git Commands 2700X i9-7980XE OpenBenchmarking.org Seconds, Fewer Is Better Git Time To Complete Common Git Commands Stock Tuned 1.2173 2.4346 3.6519 4.8692 6.0865 SE +/- 0.05, N = 3 SE +/- 0.05, N = 3 SE +/- 0.02, N = 3 SE +/- 0.02, N = 3 5.32 5.32 5.40 5.41 1. git version 2.19.1
Flexible IO Tester Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory 2700X i9-7980XE OpenBenchmarking.org MB/s, More Is Better Flexible IO Tester 3.1 Type: Sequential Write - IO Engine: Linux AIO - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test Directory Stock Tuned 1000 2000 3000 4000 5000 SE +/- 47.35, N = 12 SE +/- 93.42, N = 3 SE +/- 48.51, N = 3 SE +/- 69.94, N = 3 4796 4792 4749 4726 1. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -ll -lnuma -libverbs -lrt -laio -lz -lm -lpthread -ldl
ET: Legacy Renderer: Renderer2 - Resolution: 2560 x 1440 OpenBenchmarking.org Frames Per Second, More Is Better ET: Legacy 2.75 Renderer: Renderer2 - Resolution: 2560 x 1440 Stock Tuned 50 100 150 200 250 SE +/- 0.68, N = 3 SE +/- 1.12, N = 3 211 209
OpenArena Resolution: 2560 x 1440 OpenBenchmarking.org Frames Per Second, More Is Better OpenArena 0.8.8 Resolution: 2560 x 1440 Stock Tuned 60 120 180 240 300 SE +/- 0.42, N = 3 SE +/- 0.50, N = 3 271 269 MIN: 1 MIN: 1
ParaView Test: Wavelet Volume - Resolution: 2560 x 1440 OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Volume - Resolution: 2560 x 1440 Tuned Stock 30 60 90 120 150 SE +/- 0.61, N = 3 SE +/- 0.23, N = 3 150 149
Tesseract Resolution: 2560 x 1440 OpenBenchmarking.org Frames Per Second, More Is Better Tesseract 2014-05-12 Resolution: 2560 x 1440 Stock Tuned 70 140 210 280 350 SE +/- 5.08, N = 3 SE +/- 1.40, N = 3 341 339
R Benchmark OpenBenchmarking.org Seconds, Fewer Is Better R Benchmark Tuned Stock 0.0659 0.1318 0.1977 0.2636 0.3295 SE +/- 0.0011, N = 3 SE +/- 0.0010, N = 3 0.2915 0.2928 1. R scripting front-end version 3.5.1 (2018-07-02)
ParaView Test: Wavelet Volume - Resolution: 2560 x 1440 OpenBenchmarking.org MiVoxels / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Volume - Resolution: 2560 x 1440 Tuned Stock 500 1000 1500 2000 2500 SE +/- 9.73, N = 3 SE +/- 3.67, N = 3 2397 2391
ParaView Test: Many Spheres - Resolution: 2560 x 1440 OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.4.1 Test: Many Spheres - Resolution: 2560 x 1440 Tuned Stock 700 1400 2100 2800 3500 SE +/- 2.62, N = 3 SE +/- 1.23, N = 3 3314 3306
ParaView Test: Wavelet Contour - Resolution: 2560 x 1440 OpenBenchmarking.org MiPolys / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Contour - Resolution: 2560 x 1440 Stock Tuned 300 600 900 1200 1500 SE +/- 0.92, N = 3 SE +/- 2.43, N = 3 1387 1384
ParaView Test: Many Spheres - Resolution: 2560 x 1440 OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.4.1 Test: Many Spheres - Resolution: 2560 x 1440 Tuned Stock 8 16 24 32 40 SE +/- 0.03, N = 3 SE +/- 0.01, N = 3 33.05 32.98
Cython benchmark OpenBenchmarking.org Seconds, Fewer Is Better Cython benchmark 0.27 Stock Tuned 5 10 15 20 25 SE +/- 0.11, N = 3 SE +/- 0.03, N = 3 20.43 20.45
GLmark2 Resolution: 2560 x 1440 OpenBenchmarking.org Score, More Is Better GLmark2 276 Resolution: 2560 x 1440 Tuned Stock 600 1200 1800 2400 3000 2732 2730
ParaView Test: Wavelet Contour - Resolution: 2560 x 1440 OpenBenchmarking.org Frames / Sec, More Is Better ParaView 5.4.1 Test: Wavelet Contour - Resolution: 2560 x 1440 Stock Tuned 30 60 90 120 150 SE +/- 0.09, N = 3 SE +/- 0.23, N = 3 133 133
Sunflow Rendering System Global Illumination + Image Synthesis i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Sunflow Rendering System 0.07.2 Global Illumination + Image Synthesis Tuned Stock 0.3285 0.657 0.9855 1.314 1.6425 SE +/- 0.02, N = 12 SE +/- 0.03, N = 12 SE +/- 0.02, N = 12 SE +/- 0.02, N = 12 0.98 1.01 1.45 1.46
Apache Siege Concurrent Users: 250 i9-7980XE 2700X OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 250 Stock Tuned 10K 20K 30K 40K 50K SE +/- 642.74, N = 3 SE +/- 357.44, N = 3 SE +/- 1513.51, N = 12 SE +/- 1179.49, N = 12 46175 40011 34231 34025 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Apache Siege Concurrent Users: 200 i9-7980XE 2700X OpenBenchmarking.org Transactions Per Second, More Is Better Apache Siege 2.4.29 Concurrent Users: 200 Stock Tuned 8K 16K 24K 32K 40K SE +/- 1553.27, N = 12 SE +/- 338.75, N = 3 SE +/- 1225.15, N = 9 SE +/- 118.48, N = 3 38182 36038 31160 24403 1. (CC) gcc options: -O2 -lpthread -ldl -lssl -lcrypto
Stress-NG Test: Context Switching i9-7980XE 2700X OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Context Switching Stock Tuned 4M 8M 12M 16M 20M SE +/- 46544.97, N = 3 SE +/- 32488.23, N = 3 SE +/- 141705.02, N = 12 SE +/- 8151.19, N = 3 20723337 20720973 3880010 3700685 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Stress-NG Test: Socket Activity i9-7980XE 2700X OpenBenchmarking.org Bogo Ops/s, More Is Better Stress-NG 0.07.26 Test: Socket Activity Tuned Stock 3K 6K 9K 12K 15K SE +/- 47.31, N = 3 SE +/- 463.56, N = 11 SE +/- 22.70, N = 3 SE +/- 48.98, N = 3 15637 15571 7715 7399 1. (CC) gcc options: -O2 -std=gnu99 -lm -lz -lcrypt -lrt -lpthread -laio -lc
Hackbench Count: 8 - Type: Process i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Process Stock Tuned 4 8 12 16 20 SE +/- 0.18, N = 12 SE +/- 0.83, N = 12 SE +/- 0.07, N = 3 SE +/- 0.75, N = 12 12.23 13.58 16.48 18.13 1. (CC) gcc options: -lpthread
Hackbench Count: 4 - Type: Process i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Process Stock Tuned 3 6 9 12 15 SE +/- 0.01, N = 3 SE +/- 0.01, N = 3 SE +/- 0.50, N = 12 SE +/- 0.36, N = 12 6.64 6.65 10.96 11.45 1. (CC) gcc options: -lpthread
Hackbench Count: 16 - Type: Thread i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 16 - Type: Thread Tuned Stock 9 18 27 36 45 SE +/- 1.79, N = 12 SE +/- 1.72, N = 12 SE +/- 0.53, N = 12 SE +/- 0.70, N = 12 35.16 36.36 37.53 38.75 1. (CC) gcc options: -lpthread
Hackbench Count: 8 - Type: Thread i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 8 - Type: Thread Stock Tuned 6 12 18 24 30 SE +/- 0.38, N = 3 SE +/- 0.25, N = 3 SE +/- 0.78, N = 12 SE +/- 0.21, N = 3 20.11 20.79 23.57 25.45 1. (CC) gcc options: -lpthread
Hackbench Count: 4 - Type: Thread i9-7980XE 2700X OpenBenchmarking.org Seconds, Fewer Is Better Hackbench Count: 4 - Type: Thread Tuned Stock 4 8 12 16 20 SE +/- 0.03, N = 3 SE +/- 0.03, N = 3 SE +/- 0.26, N = 12 SE +/- 0.28, N = 3 8.82 9.08 13.96 14.39 1. (CC) gcc options: -lpthread
Sockperf Test: Latency Under Load i9-7980XE 2700X OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Under Load Stock Tuned 20 40 60 80 100 SE +/- 2.10, N = 20 SE +/- 0.27, N = 5 SE +/- 11.18, N = 15 SE +/- 8.25, N = 20 36.56 39.68 106.65 106.96 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Sockperf Test: Latency Ping Pong 2700X i9-7980XE OpenBenchmarking.org usec, Fewer Is Better Sockperf 3.4 Test: Latency Ping Pong Stock Tuned 1.035 2.07 3.105 4.14 5.175 SE +/- 0.01, N = 5 SE +/- 0.03, N = 5 SE +/- 0.04, N = 5 SE +/- 0.07, N = 20 4.20 4.25 4.40 4.60 1. (CXX) g++ options: --param -O3 -rdynamic -ldl -lpthread
Compile Bench Test: Compile i9-7980XE 2700X OpenBenchmarking.org MB/s, More Is Better Compile Bench 0.6 Test: Compile Tuned Stock 300 600 900 1200 1500 SE +/- 8.09, N = 3 SE +/- 21.07, N = 5 SE +/- 30.54, N = 12 SE +/- 17.95, N = 12 1525 1466 965 916
FS-Mark Test: 4000 Files, 32 Sub Dirs, 1MB Size 2700X i9-7980XE OpenBenchmarking.org Files/s, More Is Better FS-Mark 3.3 Test: 4000 Files, 32 Sub Dirs, 1MB Size Stock Tuned 30 60 90 120 150 SE +/- 20.85, N = 9 SE +/- 4.71, N = 12 SE +/- 5.56, N = 12 SE +/- 5.62, N = 12 131 103 113 113 1. (CC) gcc options: -static
Phoronix Test Suite v10.8.4