Differences
This shows you the differences between two versions of the page.
| Both sides previous revision Previous revision Next revision | Previous revision | ||
| doku:rstat [2015/04/28 12:12] – [Example] ir | doku:rstat [2021/09/29 12:25] (current) – [Example] goldenberg | ||
|---|---|---|---|
| Line 3: | Line 3: | ||
| [[http:// | [[http:// | ||
| - | We installed the libraries Rmpi, doMPI and foreach and their dependencies on VSC2 and VSC3. | + | We installed the libraries Rmpi, doMPI and foreach and their dependencies on VSC2 and VSC4. |
| These libraries give you the possibility to parallelize loops over more nodes with MPI using the foreach function which is very similar to a for loop. | These libraries give you the possibility to parallelize loops over more nodes with MPI using the foreach function which is very similar to a for loop. | ||
| Line 25: | Line 25: | ||
| === Parallel execution === | === Parallel execution === | ||
| - | The R-code may be parallelized in the following form: | + | In R, the code **berk-rmpi.R** |
| - | * berk-rmpi.R | + | |
| < | < | ||
| # basic example with foreach | # basic example with foreach | ||
| Line 43: | Line 42: | ||
| </ | </ | ||
| - | On VSC2, a batch job is submitted by the using the following script: | ||
| - | < | ||
| - | #$ -N rstat | ||
| - | #$ -V | ||
| - | #$ -pe mpich 16 | ||
| - | #$ -l h_rt=01: | ||
| - | echo $NSLOTS | ||
| - | mpirun -machinefile $TMPDIR/ | ||
| - | </ | ||
| - | yielding to an execution time [s] of | ||
| - | < | ||
| - | > proc.time() | ||
| - | | ||
| - | 8.495 | ||
| - | </ | ||
| On VSC3 the script reads: | On VSC3 the script reads: | ||
| < | < | ||
| Line 79: | Line 63: | ||
| 4.566 | 4.566 | ||
| </ | </ | ||
| - | |||