Job queues

Standard queues

NameCPU CPU cores per accountWalltime max of a jobPartition
cpucourtIntel Xeon Gold 6248 (2×20 cores)28072 hshared
cpulongIntel Xeon Gold 6248 (2×20 cores)8014 days (336 h)exclusive
amdAMD Epyc 7302 (2×16 cores)25696 hshared

Low priority queues

Click here to understand when to use them.

NameCPU CPU cores per accountWalltime max of a jobPartition
amdcourtAMD Epyc 7302 (2×16 cores)51236 hshared

GPU or SMP queues

Access to the GPU and/or SMP queues is only authorized for certain projects requiring GPU computing and/or nodes with high memory (after submission of a request to the scientific committee).

NameGPU cards per account CPU cores per accountWalltime max of a jobPartition
gpu4 Tesla V1003236 hshared
smp7214 days (336 h)shared

Information

The amount of CPU cores per account corresponds to the maximum number of cores that members of a same Slurm account can use simultaneously. Beyond this limit, jobs are pending.

Shared or exclusive partition:  

  • Shared: you reserve a certain number of cores on one or more nodes. Other jobs than yours can run on the remaining cores of this node(s).
  • Exclusive: you reserve all the cores on one or several nodes. No other job will run at the same time as yours on these node(s). However, even if your job does not use all the reserved cores, Slurm considers your job has consumed the elapsed time of the job multiplied by the total amount of reserved cores.

Walltime: corresponds to the maximum duration of your job before it is automatically stopped by Slurm.

Priority among the AMD nodes: see this page.

Checkpoints strongly recommended for jobs, especially long jobs. For example, see DMTCP.

Specific targeted campaigns (major challenges, jobs longer than a week, etc.) may take place at certain times of the year, for certain users or on ad hoc request.