SLURM Configration Tool

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

After you have filled in the fields of interest, use the "Submit" button on the bottom of the page to build the slurm.conf file. It will appear on your web browser. Save the file in text format as slurm.conf for use by SLURM.

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

Privacy and legal notice

Control Machines

Define the hostname of the computer on which the SLURM controller and optional backup controller will execute. You can also specify addresses of these computers if desired (defaults to their hostnames). The IP addresses can be either numeric IP addresses or names. Hostname values should should not be the fully qualified domain name (e.g. use linux rather than linux.llnl.gov).

ControlMachine: Master Controller Hostname

ControlAddr: Master Controller Address (optional)

BackupController: Backup Controller Hostname (optional)

BackupAddr: Backup Controller Address (optional)

Compute Machines

Define the machines on which user applications can run. You can also specify addresses of these computers if desired (defaults to their hostnames). Only a few of the possible parameters associated with the nodes will be set by this tool, but many others are available. All of the nodes will be placed into a single partition (or queue) with global access. Many options are availble to group nodes into partitions with a wide variety of configuration parameters. Manually edit the slurm.conf produced to exercise these options. Node names and addresses may be specified using a numeric range specification.

NodeName: Compute nodes

NodeAddr: Compute node addresses (optional)

Procs: Count of processors on each compute node

PartitionName: Name of the one partition to be created

MaxTime: Maximum time limit of jobs in minutes or INFINITE

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.

SlurmUser

Group ID Caching

If you have a slow NIS environment, big parallel jobs take a long time to start up (and may eventually time-out) because the NIS server(s) may not be able to quickly respond to simultaneous requests from multiple slurmd's. You can instruct slurmd to cache /etc/groups entries to prevent this from happening by setting CacheGroups=1. Reconfiguring ("scontrol reconfig") with CacheGroups=0 will cause slurmd to purge the cache. Select one value for CacheGroups:
0: for normal environment.
1: for slow NIS environment.

WARNING: The group ID cache does not try to keep itself in sync with the system. You MUST run "scontrol reconfig" to update the cache after making any changes to system password or group databases.

SLURM Port Numbers

The SLURM controller (slurmctld) requires a unique port for communcations as do the SLURM compute node deamonds (slurmd). If not set, slurm ports are set by checking for an entry in /etc/services and if that fails by using an interal default set at SLURM build time.

SlurmctldPort

SlurmdPort

Authentication

Define the method used for authenticating communicating between SLURM components.
Select one value for AuthType:
None: No authentication, not recommended production use
Authd: Brent Chun's authd
Munge: LLNL's Munge

Define the location of public and private SSL keys used by SLURM. These need to be generated by the SLURM administrator. Specify fully qualified pathnames. Both values are required.i

JobCredentialPrivateKey

JobCredentialPublicCertificate

State Preservation

Define the location of a directory where the slurmctld daemon saves its state. This should be a fully qualified pathname which can be read and written to by the SLURM user on both the control machine and backup controller (if configured). The location of a directory where slurmd saves state should also be defined. This must be a unique directory on each compute server (local disk).

StateSaveLocation: Slurmctld state save directory

SlurmdSpoolDir: Slurmd state save directory

Define when a DOWN node is returned to service.
Select one value for ReturnToService:
1: Automatically, when slurmd daemon registers with valid configuration
0: When explicitly restored to service by an administrator.

Scheduling

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

SchedulerAuth: authentication token (used by Wiki only)

SchedulerPort: scheduler communcations port (used by Wiki only)

Define what node configuration should be used. Using values defined in the configuration file will provide faster scheduling.
Select one value for FastSchedule:
1: Use node configuration values defined in configuration file
0: Use node configuration values actually found on each node

Interconnect

Define the node interconnect used.
Select one value for SwitchType:
Elan: Quadrics Elan3 or Elan4
Federation: IBM Federation Switch
None: No special handling required (InfiniBand, Myrinet, Ethernet, etc.)

Default MPI Type

Specify the type of MPI to be used by default. SLURM will configure environment variables accordingly. Users can over-ride this specification with an srun option.
Select one value for MpiDefault:
Mpich-Gm
MVAPICH
None: This works for most other MPI types including LAM MPI and Open MPI

Process Tracking

Define the algorithm used to identify which processes are associated with a given job. This is used signal, kill, and account for the processes associated with a job step.
Select one value for ProctrackType:
AIX: Use AIX kernel extension, recommended for AIX systems
Pgid: Use Unix Process Group ID, processes changing their process group ID can escape from SLURM control
LinuxProc: Use parent process ID records, required for MPICH-GM use, processes can escape from SLURM control
RMS: Use Quadrics kernel infrastructure, recommended for systems where this is available

Resource Selection

Define resource (node) selection algorithm to be used.
Select one value for SelectType:
Cons_res: Allocate individual processors and memory
Linear: Node-base resource allocation, does not manage indivual processor allocation
BlueGene: For IBM Blue Gene systems only

Event Logging

Slurmctld and slurmd daemons can each be configured with different levels of logging verbosity from 0 (quiet) to 7 (extremely verbose). Each may also be configured to use debug files. Use fully qualified pathnames for the files.

SlurmctldDebug (0 to 7)

SlurmctldLogFile (default is none, log goes to syslog)

SlurmdDebug (0 to 7)

SlurmdLogFile (default is none, log goes to syslog, string "%h" in name gets replaced with hostname)

Job Completion Logging

Define the job completion logging mechanism to be used.
Select one value for JobCompType:
None: No job completion logging
FileTxt: Write job completion status to a text file
Script: Use an arbitrary script to log job completion

JobCompLoc: Location specification. This is the location of the text file to be written to or the script to be run (depends upon logging mode). Use a fully qualified pathname.

Job Accounting

SLURM accounts for resource use per job. System specifics can be polled determined by system type
Select one value for JobAcctType:
None: No job accounting
Linux: Specifc Linux proc table information gathered, use with Linux systems only
JobAcctFrequency: polling interval in seconds.
JobAcctLogFile: Location specification. This is the location of the text file to be written to (used by Log only). Use a fully qualified pathname.

Process ID Logging

Define the location into which we can record the daemon's process ID. This is used for locate the appropriate daemon for signalling. Specify a specify the fully qualified pathname for the file.

SlurmctldPidFile

SlurmdPidFile

Timers

SLURM has a variety of timers to control when to consider a node DOWN, when to purge job records, how long to give a job to gracefully terminate, etc.

SlurmctldTimeout: How many seconds the backup controller waits before becoming the master controller

SlurmdTimeout: How many seconds the SLURM controller waits for the slurmd to respond to a request before considering the node DOWN

InactiveLimit: How many seconds the SLURM controller waits for srun commands to respond before considering the job or job step inactive and terminating it. A value of zero indicates unlimited wait

MinJobAge: How many seconds the SLURM controller waits after a job terminates before purging its record. A record of the job will persist in job completion and/or accounting records indefinitely, but will no longer be visible with the squeue command after puring

KillWait: How many seconds a job is given to gracefully terminate after reaching its time limit and being sent SIGTERM before sending a SIGKILLL

WaitTime: How many seconds after a job step's first task terminates before terminating all remaining tasks. A value of zero indicates unlimited wait