This shows you the differences between two versions of the page.
Next revision | Previous revision | ||
scheduling [2014/04/24 08:35] root created |
scheduling [2020/03/11 17:13] (current) root |
||
---|---|---|---|
Line 1: | Line 1: | ||
====== Scheduling ====== | ====== Scheduling ====== | ||
- | The cluster is running SLURM' | + | The cluster is running SLURM' |
- | * first-in first-out queue. | + | * "Fair share" job priorities. |
* Simple round-robin node selection. | * Simple round-robin node selection. | ||
- | The scheduler does not check which nodes are busy and try to avoid them. | + | The "Fair Share" algorithm assigns |
- | + | ||
- | This has an advantage in that it tends to leave some nodes empty for people who want a whole node. | + | |
- | + | ||
- | You can use the **-w** option | + | |
- | + | ||
- | You can use the **-x** option to avoid specified nodes. | + | |
- | + | ||
- | You can use **-c 16** to request all cores on a (standard) node. | + | |
- | + | ||
- | You can use the **--exclusive** option to as for exclusive access to all the nodes your job is allocated. This is especially useful if you have a program | + | |
+ | The scheduler does not check which nodes are busy and try to avoid them. This has an advantage in that it tends to leave some nodes empty for people who need a whole node. | ||