SLURM Configration

This form can be used to create a SLURM configuration file with you controlling many of the important configuration parameters. Many other parameters will be set to default values, but you can manually edit the resulting slurm.conf as desired for greater flexibiilty.

For more information about SLURM, see http://www.llnl.gov/linux/slurm

SLURM User

The SLURM controller (slurmctld) can run without elevated privileges, so it is recommended that a user "slurm" be created for it. For testing purposes any user name can be used.

SLURM user name

SLURM Port Numbers

The SLURM controller (slurmctld) requires a unique port for communcations as do the SLURM compute node deamonds (slurmd).

Slurmctld Port

Slurmd Port

Authentication Method

Define the method used for communicating between SLURM components. Select one:
No authentication, not recommended production use
Brent Chun's authd
LLNL's munge

Scheduler

Define the mechanism to be used for controlling job ordering. Select one:
First-In First-Out (FIFO)
FIFO with backfill
Wiki interface to Maui (configuration parameters SchedulerAuth and SchedulerPort must be manually set)

Interconnect

Define the node interconnect used. Select one:
Quadrics Elan3 and Elan4
IBM Federation Switch
No special handling required (InfiniBand, Myrinet, Ethernet, etc.)

Resource Selection

Define resource (node) selection to be used. Select one:
Allocate individual processors and memory
Node-base resource allocation, does not manage indivual processor allocation
BlueGene systems only

Head node as compute node

For smaller clusters, some admins may wish to use the head node (OSCAR server) as a compute node too. This is not recommended for larger clusters.

Run SLURM client (slurmd) on head node