System partitions

System partitions

To determine which partitions (queues) exist on a cluster you can use the sinfo -s command. Below is the command output from Cyclone.

sinfo -s
PARTITION AVAIL  TIMELIMIT   NODES(A/I/O/T) NODELIST
milan        up 1-00:00:00       16/16/2/34 ne[01-34]
skylake      up 1-00:00:00          0/3/0/3 dis[02-04]
nehalem      up 1-00:00:00          0/3/0/3 n[019-021]
cpu*         up 1-00:00:00        11/2/0/13 cn[05-17]
p100         up 1-00:00:00          1/7/0/8 cyc[01-08]
a100         up 1-00:00:00          2/4/0/6 sim[01-06]
gpu          up 1-00:00:00        16/0/0/16 gpu[01-16]

The command shows the partition name, status, maximum time limit, node availability and node list.

  • cpu: default Cyclone CPU nodes partition
  • gpu: Cyclone GPU nvidia v100 nodes partition
  • nehalem: Cytera CPU nodes partition
  • a100: Simea GPU nvidia a100 nodes partition
  • skylake: Disarm nodes partition
  • milan: AMD milan nodes partition
  • p100: Cyclamen GPU nvidia p100 nodes partition

Only the cpu and gpu partitions are used for Preparatory and Production access