Nodes are going to AU state after running job with GPU card

[Sohag Biswas]

Dear cp2k users,

Hello to all. I need your help. when i am submitting cp2k job all the nodes
in the cluster are going out of memory after almost 70% jobs are completion
stage. Are any temporary files are created during the execution process, if
then how will they vary? What must be scratch, swap and root partition
sizes for good environment of cp2k execution?

Thank you very much.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20150419/0bef04fd/attachment.htm>