Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| infra:compute:slurmconfig [2025/02/18 18:37] – remi | infra:compute:slurmconfig [2025/11/10 12:48] (current) – [User creation] ask the staff, not Rémi marc | ||
|---|---|---|---|
| Line 43: | Line 43: | ||
| These partitions can be used to restrain an account to use only one server. | These partitions can be used to restrain an account to use only one server. | ||
| + | More details of partitions usage TODO | ||
| ===== Accounts ===== | ===== Accounts ===== | ||
| Line 54: | Line 54: | ||
| There are 2 other groups : Test and temp : Test is only for administration purpose, and temp is a locked group either to migrate someone from another account (can't delete an account when someone has it as a default account) or to disallow someone to run jobs (MaxSubmitJob=0) | There are 2 other groups : Test and temp : Test is only for administration purpose, and temp is a locked group either to migrate someone from another account (can't delete an account when someone has it as a default account) or to disallow someone to run jobs (MaxSubmitJob=0) | ||
| + | TODO | ||
| Line 66: | Line 66: | ||
| * MaxTRES=gres/ | * MaxTRES=gres/ | ||
| * GrpTRES=gres/ | * GrpTRES=gres/ | ||
| - | * GrpWall=08:00:00 | + | * GrpWall=3-00:00:00 |
| - | * MaxWall=08:00:00 | + | * MaxWall=3-00:00:00 |
| * **standard_rs :** | * **standard_rs :** | ||
| * MaxCPUs=24 | * MaxCPUs=24 | ||
| - | * MaxNodes=2 | + | * MaxNodes=1 |
| * MaxTRES=gres/ | * MaxTRES=gres/ | ||
| * GrpTRES=gres/ | * GrpTRES=gres/ | ||
| - | * GrpWall=04:00:00 | + | * GrpWall=1-00:00:00 |
| - | * MaxWall=04:00:00 | + | * MaxWall=1-00:00:00 |
| + | **TODO :** complete / modify until limits are finished | ||
| ===== Scheduling ===== | ===== Scheduling ===== | ||
| Line 86: | Line 87: | ||
| * Fairshare : 250 | * Fairshare : 250 | ||
| + | Then on each account, everyone has a 100 fairshare : SLURM is just making calculation then on the percentage using both the share in each group, and the parent account premium / standard. | ||
| Line 95: | Line 96: | ||
| Several users can share the same project account to work as a team. (Limits are applied both to each user for maximum limits and the group for group limits) | Several users can share the same project account to work as a team. (Limits are applied both to each user for maximum limits and the group for group limits) | ||
| - | Currently, Rémi creates and configure user accounts. | + | The [[infra: |
| + | |||
| + | [[infra: | ||
| + | |||
| + | |||
| + | |||
| + | ===== Backups ===== | ||
| + | |||
| + | ==== Cluster configuration ==== | ||
| + | |||
| + | Currently, all the SLURM configuration is manually backuped (files, DB) | ||
| + | |||
| + | TODO : automate and redirect to backup server when its ready | ||
| + | |||
| + | ==== User data ==== | ||
| + | |||
| + | For user data, currently compute users have to request some space on the filer01.hevs.ch server from the Sinf : there is a [[ https:// | ||
| + | Consider all space on the Compute center to be short-lived, | ||