I would say so.
Certainly, if you have many nodes and/or many jobs being submitted, you
will see an impact, but in my experience comparing Slurm to SGE, Slurm
has much less overhead to cause as much impact.
Brian Andrus
On 2/26/2020 1:05 PM, Joshua Baker-LePain wrote:
On Wed, 26 Feb 2020 at 12:56pm, Brian Andrus wrote
Any shared filesystem that both systems can get to will work.
I have done it with NFS, Gluster, appliances (NetApp), etc.
Being in a separate datacenter is fine, but you will see some
latency, which you likely already addressed if you are pysically
splitting a network like that.
Sure, NFS would be the default. I was just worried about the
performance. We tried putting the queue directory on NFS with our SGE
shadow master config, and definitely saw a job throughput/scheduling
performance hit. Would the hit be less with slurm?