[Berzelius-users] Berzelius downtime on 2nd of May

[Filip Polbratt]

Dear Berzelius Users,

All maintenance on the login nodes and scheduler was completed by 13:15. 
Maintenance of the thin compute nodes (DGX A100 with 40GB GPUs) has also 
been completed. The fat compute nodes (DGX A100 with 80GB GPUs) will 
remain unavailable over night for additional testing.

Best regards,
Filip Polbratt

On 2024-04-22 10:24, Filip Polbratt wrote:
> Dear Berzelius Users,
> 
> on Thursday the 2nd of May we will start a maintenance window for 
> several tasks that require downtime of the cluster. We are scheduling 
> this maintenance window to start on the 2nd at 08:00 and to last the 
> entire day. The primary reason for the downtime is to update our 
> scheduler Slurm. This is a low risk routine operation. No external staff 
> are involved.
> 
> *A* login node will be available at all times, but all login nodes will 
> be rebooted during the maintenance. Use berzelius.nsc.liu.se to connect 
> instead of connecting directly to a login node. Interactive tasks 
> running on the nodes will be stopped and must be restarted manually 
> after the maintenance window. Some compute nodes will likely be returned 
> to service early.
> 
> Best regards,
> Filip Polbratt
> NSC
> _______________________________________________
> Berzelius-users mailing list
> Berzelius-users at lists.nsc.liu.se
> https://lists.nsc.liu.se/mailman/listinfo/berzelius-users