Dear Users,
May I have your guidance?
How to run the multiple job in the server,
We have 2 servers Platinum and Cerium,
1. when I launch the 2 job in Platinum the tool launches successfully and
distribute the job to 2 different servers. but while launching the 3rd job the
resource is in queu
Hi,
We are using slurm 21.08. We are curious to know how to use "sbank" utility
for crediting GPU Hours , just like cpu minutes, and also get the status of
GPUHours credited, used etc.
Actually, sbank utility from github is not having functionality of adding /
querying the GPUHours
Any other mean
On 10-12-2023 17:29, Ryan Novosielski wrote:
This is basically always somebody filling up /tmp and /tmp residing on the same
filesystem as the actual SlurmdSpoolDirectory.
/tmp, without modifications, it’s almost certainly the wrong place for
temporary HPC files. Too large.
Agreed! That's w
We maintain /tmp as a separate partition to mitigate this exact scenario on all
nodes though it doesn’t necessarily need to be part of the primary system RAID.
No need for tmp resiliency.
Regards,
Peter
Peter Goode
Research Computing Systems Administrator
Lafayette College
> On Dec 10, 2023,
This is basically always somebody filling up /tmp and /tmp residing on the same
filesystem as the actual SlurmdSpoolDirectory.
/tmp, without modifications, it’s almost certainly the wrong place for
temporary HPC files. Too large.
Sent from my iPhone
> On Dec 8, 2023, at 10:02, Xaver Stiensmeie
Hello Brian Andrus,
we ran 'df -h' to determine the amount of free space I mentioned below.
I also should add that at the time we inspected the node, there was
still around 38 GB of space left - however, we were unable to watch the
remaining space while the error occurred so maybe the large file(
