Différences
Ci-dessous, les différences entre deux révisions de la page.
| Les deux révisions précédentes Révision précédente Prochaine révision | Révision précédente | ||
| bench_mkl_lapack [2016/10/21 12:00] – admin | bench_mkl_lapack [2016/10/21 12:35] (Version actuelle) – [Compile with LAPACK (and OpenBLAS)] admin | ||
|---|---|---|---|
| Ligne 1: | Ligne 1: | ||
| ====== MKL vs LAPACK ====== | ====== MKL vs LAPACK ====== | ||
| - | We strongly encourage the use of MKL instead of Lapack because | + | We strongly encourage the use of Intel MKL instead of Lapack |
| MKL is more efficient and optimized for the CPU architecture. | MKL is more efficient and optimized for the CPU architecture. | ||
| + | < | ||
| ===== Benchmarks ===== | ===== Benchmarks ===== | ||
| Ligne 59: | Ligne 59: | ||
| </ | </ | ||
| - | ==== Compile with Lapack | + | ==== Compile with LAPACK |
| <code bash> | <code bash> | ||
| $ module load gcc | $ module load gcc | ||
| - | $gfortran -O2 randomsys.f90 -o randomsys_lapack.exe -lblas -llapack | + | $ gfortran -O2 randomsys.f90 -o randomsys_lapack.exe -lblas -llapack |
| </ | </ | ||
| + | |||
| Ligne 73: | Ligne 74: | ||
| $ ifort -O2 -mkl randomsys.f90 -o randomsys_mkl.exe | $ ifort -O2 -mkl randomsys.f90 -o randomsys_mkl.exe | ||
| </ | </ | ||
| + | |||
| + | |||
| + | |||
| + | ==== Results ==== | ||
| + | |||
| + | |N|LAPACK|MKL (1 core) | MKL (16 core) | | ||
| + | |4000|14s|6s|2s| | ||
| + | |6000|41s|12s|4s| | ||
| + | |8000|1m27s|25s|5s| | ||
| + | |10000|**2mn46s**|**45s**|**8s**| | ||
| + | |||
| + | <note tip>For MKL, use '' | ||