diff --git a/doc/man/man5/slurm.conf.5 b/doc/man/man5/slurm.conf.5 index 80591dcf0cd14297ce09ee4df300ef36d5e763e9..a79bc09225eefaa3c5769972ef1f427f891bbf1f 100644 --- a/doc/man/man5/slurm.conf.5 +++ b/doc/man/man5/slurm.conf.5 @@ -2662,12 +2662,16 @@ Also see the partition configuration parameter \fBLLN\fR use the least loaded nodes in selected partitions. .TP \fBCR_Pack_Nodes\fR -Rather than evenly distributing a job step's tasks evenly across it's allocated -nodes, pack them as tightly as possible on the nodes. For example, a job step -starting ten tasks acrosss a job allocation consisting of two full nodes each -with eight CPUs (i.e. the job allocation includes 16 CPUs), without this option -will start five tasks on each of the two nodes. With this option, eight tasks -will be started on the first node and two on the second node. +If a job allocation contains more resources than will be used for launching +tasks (e.g. if whole nodes are allocated to a job), then rather than evenly +distributing a job's tasks evenly across it's allocated nodes, pack them as +tightly as possible on thsee nodes. +For example, consider a job allocation containing two \fBentire\fR nodes with +eight CPUs each. +If the job starts ten tasks across those two nodes without this option, it will +start five tasks on each of the two nodes. +With this option, eight tasks will be started on the first node and two tasks +on the second node. .TP \fBCR_Socket\fR Sockets are consumable resources.