...
- At the moment there are no time limits for the idle-cycle jobsGPAW jobs will suspend/resume automatically when other higher-priority jobs start/finish
- Different codes work on different systems. Contact vossj for details. The exception is dacapo/jacapo, which does not suspend correctly
- We are using other people's computing resources, so it is important to try not to do things like use up so much memory that the machines will crash
- I believe psnehidle and pinto have fast infiniband interconnects, which may be useful for vdW/BEEF jobs that are larger than 1 node
Cluster | Submit Command | Queue Name(s) | Cores per Node | Memory per Node | Current User Interconnect |
---|---|---|---|---|---|
pinto | gpaw-pinto-ver-bsub | idle | 8 | 48GB | brogaard Infiniband |
oak | gpaw-oak-ver-bsub | idle | 4 | 16GB | 1Gb ethernet |
fell | esp-fell-ver-bsub | idle | 8 | 16GB monicag2 | 1Gb ethernet |
lcls | gpawesp-lcls-ver-bsub | psfehidle/psnehidle/psfehidle | 12 | 24GB | Infiniband |
clustered systems | standard commands | psanacs-idleq | 16 | 32GB | 1Gb ethernet jvarley |