This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
cluster:134 [2014/08/20 10:29] hmeij [High Throughput] |
cluster:134 [2014/08/21 16:00] hmeij |
||
---|---|---|---|
Line 101: | Line 101: | ||
[[https:// | [[https:// | ||
+ | |||
+ | Vanilla out of the box with these changes | ||
* MaxJobCount=120000 | * MaxJobCount=120000 | ||
Line 163: | Line 165: | ||
^NrJobs^N^hh: | ^NrJobs^N^hh: | ||
- | | 1,000|8|00:??:??| | + | | 1,000|8|00:05:00| |
- | | 5,000|8|00:??:??| | + | | 5,000|8|00:23:43| |
- | |10, | + | |10, |
- | |25, | + | |25, |
+ | |||
+ | ==== MPI ==== | ||
+ | |||
+ | With sbatch there is no need for a wrapper script, slurm figures it all out. | ||
+ | |||
+ | < | ||
+ | |||
+ | #!/bin/bash | ||
+ | #/ | ||
+ | |||
+ | #SBATCH --job-name=" | ||
+ | #SBATCH --ntasks=8 | ||
+ | #SBATCH --begin=now | ||
+ | |||
+ | # unique job scratch dir | ||
+ | #export MYLOCALSCRATCH=/ | ||
+ | #cd $MYLOCALSCRATCH | ||
+ | |||
+ | echo " | ||
+ | |||
+ | rm -rf err out logfile mdout restrt mdinfo | ||
+ | |||
+ | export PATH=/ | ||
+ | export LD_LIBRARY_PATH=/ | ||
+ | which mpirun | ||
+ | |||
+ | mpirun / | ||
+ | |||
+ | #/ | ||
+ | |||
+ | </ | ||
\\ | \\ | ||
**[[cluster: | **[[cluster: |