kernel built with -march=native, no other tweaks, powerdxx running. Intel exposes all cores for frequency
Compare your own system(s) to this result file with the
Phoronix Test Suite by running the command:
phoronix-test-suite benchmark 2302022-CFOR-230201970 m610-vs-m915-memory - Phoronix Test Suite m610-vs-m915-memory kernel built with -march=native, no other tweaks, powerdxx running. Intel exposes all cores for frequency
HTML result view exported from: https://openbenchmarking.org/result/2302022-CFOR-230201970&sor&grt .
m610-vs-m915-memory Processor Motherboard Chipset Memory Disk Graphics OS Kernel Compiler File-System M610 2x5687 96GB, Kern=native M610 2x5687 96GB, Kern=native NoPowerD but epp running Intel Xeon X5687 @ 3.59GHz (16 Cores) Dell PowerEdge M610x Intel 5500 I/O Hub ESI Port 96GB ATA OCZ VERTEX2 1.35 + DP BACKPLANE 1.07 Matrox MGA G200eW WPCM450 FreeBSD 13.1-RELEASE (x86_64) Clang 13.0.0 zfs OpenBenchmarking.org Security Details - KPTI
m610-vs-m915-memory cachebench: Read Cache cachebench: Write Cache mbw: Memory Copy - 1024 MiB mbw: Memory Copy, Fixed Block Size - 1024 MiB stream: Copy stream: Scale stream: Triad stream: Add t-test1: 1 t-test1: 2 M610 2x5687 96GB, Kern=native M610 2x5687 96GB, Kern=native NoPowerD but epp running 4761.698116 6862.703033 4443.330 4533.920 27283.5 25830.9 28376.0 27760.2 30.659 9.950 9132.155288 13329.508858 5468.251 5113.852 28200.7 26531.0 29094.6 29049.2 14.657 4.490 OpenBenchmarking.org
CacheBench Read Cache OpenBenchmarking.org MB/s, More Is Better CacheBench Read Cache M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 2K 4K 6K 8K 10K SE +/- 0.13, N = 3 SE +/- 12.44, N = 3 9132.16 4761.70 MIN: 9130.28 / MAX: 9132.98 MIN: 4296.52 / MAX: 5603.47 1. (CC) clang options: -lrt
CacheBench Write Cache OpenBenchmarking.org MB/s, More Is Better CacheBench Write Cache M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 3K 6K 9K 12K 15K SE +/- 0.04, N = 3 SE +/- 36.23, N = 3 13329.51 6862.70 MIN: 12461.51 / MAX: 13680.67 MIN: 6118.26 / MAX: 8319.42 1. (CC) clang options: -lrt
MBW Test: Memory Copy - Array Size: 1024 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy - Array Size: 1024 MiB M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 1200 2400 3600 4800 6000 SE +/- 14.78, N = 3 SE +/- 53.56, N = 4 5468.25 4443.33 1. (CC) clang options: -O3 -march=native
MBW Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB OpenBenchmarking.org MiB/s, More Is Better MBW 2018-09-08 Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiB M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 1100 2200 3300 4400 5500 SE +/- 135.10, N = 15 SE +/- 95.14, N = 12 5113.85 4533.92 1. (CC) clang options: -O3 -march=native
Stream Type: Copy OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Copy M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 6K 12K 18K 24K 30K SE +/- 27.85, N = 5 SE +/- 289.67, N = 5 28200.7 27283.5 1. (CC) clang options: -mcmodel=medium -O3 -march=native -fopenmp
Stream Type: Scale OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Scale M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 6K 12K 18K 24K 30K SE +/- 90.98, N = 5 SE +/- 331.70, N = 5 26531.0 25830.9 1. (CC) clang options: -mcmodel=medium -O3 -march=native -fopenmp
Stream Type: Triad OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Triad M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 6K 12K 18K 24K 30K SE +/- 131.42, N = 5 SE +/- 363.58, N = 5 29094.6 28376.0 1. (CC) clang options: -mcmodel=medium -O3 -march=native -fopenmp
Stream Type: Add OpenBenchmarking.org MB/s, More Is Better Stream 2013-01-17 Type: Add M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 6K 12K 18K 24K 30K SE +/- 123.58, N = 5 SE +/- 409.47, N = 5 29049.2 27760.2 1. (CC) clang options: -mcmodel=medium -O3 -march=native -fopenmp
t-test1 Threads: 1 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 1 M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 7 14 21 28 35 SE +/- 0.00, N = 3 SE +/- 0.22, N = 3 14.66 30.66 1. (CC) clang options: -pthread
t-test1 Threads: 2 OpenBenchmarking.org Seconds, Fewer Is Better t-test1 2017-01-13 Threads: 2 M610 2x5687 96GB, Kern=native NoPowerD but epp running M610 2x5687 96GB, Kern=native 3 6 9 12 15 SE +/- 0.006, N = 3 SE +/- 0.046, N = 3 4.490 9.950 1. (CC) clang options: -pthread
Phoronix Test Suite v10.8.4