Hi Loris, You might check out the Weight parameter in the Node section of the slurm.conf documentation. I believe you could just give the fat nodes a higher node weight than the thinner nodes, to achieve your goal.
Best, Lyn On Fri, Jan 31, 2014 at 3:31 AM, Loris Bennett <[email protected]>wrote: > > Hi, > > We have nodes with different amounts of memory in the same partition. > Currently we have the situation that the job with the highest priority > can only run on a fat node and has to wait because they are all full. > > Other jobs, which could run on thin nodes, but have a lower priority, > don't start. > > Is this the intended behaviour? If so, can SLURM be configured to > behave differently? > > This is observed with version 2.2.7, so things may well have changed. > > Regards > > Loris > > -- > This signature is currently under construction. >
