On 8/21/19 3:00 PM, Richard Edwards wrote:
So I am starting to see a pattern. Some combination of CentOS + Ansible + OpenHPC + SLURM + Old CUDA/Nvidia Drivers;-).
My only comment there would be I do like xCAT, especially with statelite settings so you can PXE boot a RAM disk on the nodes but still set up parts of the image that are writeable over NFS from the management node for persistent storage. Tends to help if you have IPMI on the nodes though (remote power control etc).
https://xcat.org/ Best of luck! Chris -- Chris Samuel : http://www.csamuel.org/ : Berkeley, CA, USA _______________________________________________ Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit https://beowulf.org/cgi-bin/mailman/listinfo/beowulf
