Linpack
Intelのウェブサイトでダウンロードできる。
Windows、Linux、macOS用があって、コンパイル済みのバイナリになっているのでダウンロードして実行するだけ。
root@ds220g1:/tmp/l_mklb_p_2020.2.001/benchmarks_2020/linux/mkl/benchmarks/linpack# ./runme_xeon64 This is a SAMPLE run script for running a shared-memory version of Intel(R) Distribution for LINPACK* Benchmark. Change it to reflect the correct number of CPUs/threads, problem input files, etc.. *Other names and brands may be claimed as the property of others. ./runme_xeon64: line 28: [: too many arguments Sat Oct 31 22:27:28 PDT 2020 Sample data file lininput_xeon64. Current date/time: Sat Oct 31 22:27:28 2020 CPU frequency: 3.699 GHz Number of CPUs: 2 Number of cores: 36 Number of threads: 36 Parameters are set to: Number of tests: 15 Number of equations to solve (problem size) : 1000 2000 5000 10000 15000 18000 20000 22000 25000 26000 27000 30000 35000 40000 45000 Leading dimension of array : 1000 2000 5008 10000 15000 18008 20016 22008 25000 26000 27000 30000 35000 40000 45000 Number of trials to run : 4 2 2 2 2 2 2 2 2 2 1 1 1 1 1 Data alignment value (in Kbytes) : 4 4 4 4 4 4 4 4 4 4 4 1 1 1 1 Maximum memory requested that can be used=16200901024, at the size=45000 =================== Timing linear equation system solver =================== Size LDA Align. Time(s) GFlops Residual Residual(norm) Check 1000 1000 4 0.171 3.9140 1.040126e-12 3.167791e-02 pass 1000 1000 4 0.011 61.1942 1.040126e-12 3.167791e-02 pass 1000 1000 4 0.008 85.7699 1.040126e-12 3.167791e-02 pass 1000 1000 4 0.007 99.8466 1.040126e-12 3.167791e-02 pass 2000 2000 4 0.027 200.7489 3.906653e-12 3.106923e-02 pass 2000 2000 4 0.027 198.6520 3.906653e-12 3.106923e-02 pass 5000 5008 4 0.128 651.3254 2.713202e-11 3.582731e-02 pass 5000 5008 4 0.110 758.6510 2.713202e-11 3.582731e-02 pass 10000 10000 4 1.169 570.6969 1.157969e-10 3.914694e-02 pass 10000 10000 4 0.529 1259.7720 1.157969e-10 3.914694e-02 pass 15000 15000 4 1.573 1430.3173 2.086843e-10 3.158888e-02 pass 15000 15000 4 1.569 1433.9648 2.086843e-10 3.158888e-02 pass 18000 18008 4 2.662 1460.9114 3.281892e-10 3.476062e-02 pass 18000 18008 4 2.660 1461.7944 3.281892e-10 3.476062e-02 pass 20000 20016 4 3.657 1458.7962 3.922531e-10 3.362530e-02 pass 20000 20016 4 3.660 1457.4162 3.922531e-10 3.362530e-02 pass 22000 22008 4 4.766 1489.6896 4.458339e-10 3.153157e-02 pass 22000 22008 4 4.793 1481.2429 4.458339e-10 3.153157e-02 pass 25000 25000 4 7.018 1484.4988 5.538188e-10 3.063248e-02 pass 25000 25000 4 7.023 1483.3266 5.538188e-10 3.063248e-02 pass 26000 26000 4 7.730 1516.0049 6.363453e-10 3.252047e-02 pass 26000 26000 4 7.734 1515.2853 6.363453e-10 3.252047e-02 pass 27000 27000 4 8.545 1535.7519 6.926203e-10 3.270807e-02 pass 30000 30000 1 11.645 1545.8491 9.123060e-10 3.509794e-02 pass 35000 35000 1 18.625 1534.8473 1.203062e-09 3.408724e-02 pass 40000 40000 1 27.528 1550.0281 1.566522e-09 3.345501e-02 pass 45000 45000 1 39.394 1542.2125 1.876337e-09 3.225705e-02 pass Performance Summary (GFlops) Size LDA Align. Average Maximal 1000 1000 4 62.6812 99.8466 2000 2000 4 199.7004 200.7489 5000 5008 4 704.9882 758.6510 10000 10000 4 915.2344 1259.7720 15000 15000 4 1432.1411 1433.9648 18000 18008 4 1461.3529 1461.7944 20000 20016 4 1458.1062 1458.7962 22000 22008 4 1485.4663 1489.6896 25000 25000 4 1483.9127 1484.4988 26000 26000 4 1515.6451 1516.0049 27000 27000 4 1535.7519 1535.7519 30000 30000 1 1545.8491 1545.8491 35000 35000 1 1534.8473 1534.8473 40000 40000 1 1550.0281 1550.0281 45000 45000 1 1542.2125 1542.2125 Residual checks PASSED End of tests Done: Sat Oct 31 22:30:50 PDT 2020 root@ds220g1:/tmp/l_mklb_p_2020.2.001/benchmarks_2020/linux/mkl/benchmarks/linpack#
最大で1.5TFlopsくらい性能が出ているという読み方で良いのかな?