SLURM configuration

This section details the current SLURM configuration for the ISC Computational Center.

SLURM has been installed from tarball, version : 24.11.0

All official plugins installed :

• libnvidia-ml
• TODO

Install / upgrade process TODO

• Client (slurm-smd-client)
• Worker (slurm-smd,slurm-smd-slurmd)
• Controller (slurm-smd-slurmctld)
• Accounting DB (slurm-smd-slurmdbd)

• Client (slurm-smd-client)
• Worker (slurm-smd,slurm-smd-slurmd)

TODO

This is the default partition : is is currently composed of Chacha and Disco.

These partitions can be used to restrain an account to use only one server.

Accounts have been created in 2 groups :

• Premium Researchers (premium_rs) :
• Standard Researchers (standard_rs) : All users who can't participate financially to the project. Students are also part of this group

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)

Current limits on QOS and accounts applied to each project account :

• premium_rs :
  • MaxCPUs=44
  • MaxNodes=2
  • MaxTRES=gres/gpu=1,gres/shard=96,cpu=44,mem=500G
  • GrpTRES=gres/gpu=1,gres/shard=96,cpu=44,mem=500G
  • GrpWall=08:00:00
  • MaxWall=08:00:00

• standard_rs :
  • MaxCPUs=24
  • MaxNodes=2
  • MaxTRES=gres/gpu=1,gres/shard=96,cpu=24,mem=256G
  • GrpTRES=gres/gpu=1,gres/shard=96,cpu=24,mem=256G
  • GrpWall=04:00:00
  • MaxWall=04:00:00

Fairshare is one way of priorizing jobs in the job queue.

• premium_rs :
  • Fairshare : 750
• standard_rs :
  • Fairshare : 250

TODO

Users fill their form from the page ISC Computational Center , then their information is used to create their SSH access, and their SLURM user in its project account, according to the SLA / QOS we can provide : Premium or Standard.

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.