Thanks Chris, my user says that it worked EXACTLY as he had imagined.
Sweet!!!
I love slurm.
On 10/05/16 12:25, Mark Moorcroft wrote:
> Awesome. That worked. I moved it from /tmp to /scratch, which in our
> case is the bulk of the local free space on all the drives (login and
> compute). Although, since it doesn't track free space it may be the
> wrong approach.
Well after this I set it
Awesome. That worked. I moved it from /tmp to /scratch, which in our case
is the bulk of the local free space on all the drives (login and compute).
Although, since it doesn't track free space it may be the wrong approach.
Regardless, now I just need to know how you specify a minimum TmpSpace in a
On 10/05/16 07:04, Mark Moorcroft wrote:
> I think --tmp would be something different from a dedicated partition
> for userland scratch space?
I think it depends.
Slurm only allows you to specify a single temporary filesystem to report back
on:
TmpFS Fully qualified pathname of the file syst
I think --tmp would be something different from a dedicated partition for
userland scratch space?
> On May 9, 2016, at 4:11 PM, Mark Moorcroft wrote:
>
> I got this from a cluster user...
>
> The nodes have different sized scratch partitions. I need to get some jobs
> on the nodes with more scratch space.
> I know how to asked for a specific node, but it would be better to make a
> reque