Useful Slurm Commands#
Slurm provides a variety of tools that allow a user to manage and understand their jobs. This tutorial will introduce these tools, as well as provide details on how to use them.
Finding queuing information with squeue#
The squeue command is a tool we use to pull up information about the
jobs currently in the Slurm queue. By default, the squeue command will print out the
job ID, partition, username, job status,
number of nodes, and name of nodes for all jobs queued or
running within Slurm. Usually, you wouldn’t need information for all
jobs that were queued in the system, so we can specify jobs that only
you are running with the --user flag:
$ squeue --user=your_rc-username
We can output non-abbreviated information with the --long flag. This
flag will print out the non-abbreviated default information with the
addition of a timelimit field:
$ squeue --user=your_rc-username --long
The squeue command also provides users with a means to calculate a
job’s estimated start time by adding the --start flag to our
command. This will append Slurm’s estimated start time for each job in
our output information.
Note
The start time provided by this command can be inaccurate. This is because the time calculated is based on jobs queued or running in the system. If a job with a higher priority is queued after the command is run, your job may be delayed.
$ squeue --user=your_rc-username --start
When checking the status of a job, you may want to repeatedly call the
squeue command to check for updates. We can accomplish this by adding
the --iterate flag to our squeue command. This will run squeue every
n seconds, allowing for a frequent, continuous update of queue
information without needing to repeatedly call squeue:
$ squeue --user=your_rc-username --start --iterate=n_seconds
Press ctrl-c to stop the command from looping and bring you back
to the terminal.
Important
Do not use an --iterate= value less than 60 (i.e. 1 minute). Shorter iterations can overwhelm the Slurm controller and lead to the suspension of RC accounts.
$ squeue --user=your_rc-username --start --iterate=60
See also
For more information, visit the Slurm page on squeue
Assessing the priority and queue position of a job with sprio#
The Slurm sprio command enables users to view the priority score of their pending job(s), including the factors that contribute to the job’s priority. The command can also be used to determine the job’s position in the queue. The sprio command is most useful when it is run in a manner that sorts by the priority score, since your job will be behind jobs with higher priority scores:
sprio -S '-Y'
The -S flag tells the command to sort the output by priority in descending order, -Y. The output includes the following columns, in order: jobid, partition, priority, as well as the five factors that are added together to determine priority: site, age, fairshare, jobsize, and qos. Fairshare is generally the dominant factor, which quantifies your system usage over the past 28 days relative to your expected usage in a given allocation (account). You can learn more about Fairshare in our Alpine Allocations page.
The output from the above command will show results across all partitions. You will likely be most interested in the relative priority of your job compared to all other pending jobs in a specific partition. This can be done as follows, (e.g., for the aa100 partition):
sprio -S '-Y' -p aa100
You can either scroll through the output to find your job, or you can find its numeric position in the partition queue as follows:
sprio -S '-Y' -p aa100 | grep -n 12345678
The modifier pipes the output of sprio into the linux pattern-matching command grep, and shows the line number (-n) of the jobid you specify (12345678). The line number is the job’s position in the partition queue.
Important
Please note that the above command will provide the current position of your job in the queue. The queue is constantly changing, so please take the sprio information as a helpful guide but not an absolute. Your job’s position in the queue will change over time.
See also
For more information, visit the Slurm page on sprio
Stopping jobs with scancel#
Sometimes you may need to stop a job entirely while it’s running. The
best way to accomplish this is with the scancel command. The scancel
command allows you to cancel jobs you are running on Research
Computing resources using the job’s ID.
$ scancel your_job-id
To cancel multiple jobs, you can use a comma-separated list of job IDs:
$ scancel your_job-id1, your_job-id2, your_job-id3
See also
For more information, visit the Slurm manual on scancel
Analyzing currently running jobs with sstat#
The sstat command allows users to easily pull up status information
about their currently running jobs. This includes information about CPU usage,
task information, node information, resident set size
(RSS), and virtual memory (VM). We can invoke the sstat
command as such:
$ sstat --jobs=your_job-id
The default output from sstat may not include all the information you need. To remedy this,
we can use the --format flag to choose what we want in our
output. The format flag takes a list of comma-separated variables
that specify output data:
$ sstat --jobs=your_job-id --format=var_1,var_2, ... , var_N
A chart of some of these variables is listed in the table below:
Variable |
Description |
|---|---|
avecpu |
Average CPU time of all tasks in a job. |
averss |
Average resident set size of all tasks. |
avevmsize |
Average virtual memory of all tasks in a job. |
jobid |
The id of the Job. |
maxrss |
Maximum resident set size (peak physical memory usage) across all tasks in the job. |
maxvmsize |
Maximum virtual memory size (peak memory allocated, including malloced memory) across all tasks. |
ntasks |
Number of tasks in a job. |
Note
sstat requires either a full job step ID (e.g., 123456.0) or the -a flag to return results. Supplying only the job ID (e.g., 123456) may result in no output.
As an example, let’s print out a job’s id, average cpu time, max rss, and the number of tasks. We can do this by typing out the command:
sstat --jobs=your_job-id -a --format=jobid,avecpu,maxrss,ntasks
See also
A full list of variables that specify data handled by sstat can be
found with the --helpformat flag or by visiting the slurm page on
sstat.
Analyzing past jobs with sacct#
The sacct command allows users to pull up status information about past jobs. This command is very similar to sstat, but is used on jobs that have been previously run on the system instead of currently running jobs. We can pull up accounting information on jobs based on the:
Job ID:
$ sacct --jobs=your_job-id
Research Computing Username:
$ sacct --user=your_rc-username
By default, sacct will only pull up jobs that were run on the current day. We can use the --starttime flag to tell the command to look beyond its short-term cache of jobs.
$ sacct –-jobs=your_job-id –-starttime=YYYY-MM-DD
To see a non-abbreviated version of sacct output, use the --long
flag:
$ sacct –-jobs=your_job-id –-starttime=YYYY-MM-DD --long
Formatting sacct output#
Like sstat, the standard output of sacct may not provide the
information we want. To remedy this, we can use the --format flag to
choose what we want in our output. Similarly, the format flag is
handled by a list of comma-separated variables that specify output
data:
$ sacct --user=your_rc-username --format=var_1,var_2, ... ,var_N
A chart of some of these variables is provided below:
Variable |
Description |
|---|---|
account |
Account the job ran under. |
avecpu |
Average CPU time of all tasks in the job. |
averss |
Average resident set size of all tasks in the job. |
cputime |
Formatted (Elapsed time * CPU) count used by a job or step. |
elapsed |
Jobs elapsed time formatted as |
exitcode |
The exit code returned by the job script or salloc. |
jobid |
The id of the Job. |
jobname |
The name of the Job. |
maxdiskread |
Maximum number of bytes read by all tasks in the job. |
maxdiskwrite |
Maximum number of bytes written by all tasks in the job. |
maxrss |
Maximum resident set size of all tasks in the job. |
ncpus |
Amount of allocated CPUs. |
nnodes |
The number of nodes used in a job. |
ntasks |
Number of tasks in a job. |
priority |
Slurm priority. |
qos |
Quality of service. |
reqcpu |
Required number of CPUs |
reqmem |
Required amount of memory for a job. |
user |
Username of the person who ran the job. |
As an example, suppose you want to find information about jobs that were run on March 12, 2024. You want to show information regarding the job name, the number of nodes used in the job, the number of cpus, the maxrss, and the elapsed time. Your command would look like this:
$ sacct --jobs=your_job-id --starttime=2024-03-12 --format=jobname,nnodes,ncpus,maxrss,elapsed
As another example, suppose you would like to pull up information on jobs that were run on February 21, 2024. You would like information on job ID, job name, QoS, Number of Nodes used, Number of CPUs used, Maximum RSS, CPU time, Average CPU time, and elapsed time. Your command would look like this:
$ sacct –-jobs=your_job-id –-starttime=2024-02-21 --format=jobid,jobname,qos,nnodes,ncpu,maxrss,cputime,avecpu,elapsed
See also
A full list of variables that specify data handled by sacct can be
found with the --helpformat flag or by visiting the slurm page on
sacct.
Controlling queued and running jobs using scontrol#
The scontrol command provides users extended control of their jobs
run through Slurm. This includes actions like suspending a job,
holding a job from running, or pulling extensive status information on
jobs.
To suspend a job that is currently running on the system, we can use
scontrol with the suspend command. This will stop a running job on
its current step that can be resumed at a later time. We can suspend a
job by typing the command:
$ scontrol suspend job_id
To resume a paused job, we use scontrol with the resume command:
$ scontrol resume job_id
Slurm also provides a utility to hold jobs that are queued in the
system. Holding a job will place the job in the lowest priority,
effectively “holding” the job from being run. A job can only be held
if it’s waiting on the system to be run. We use the hold command to
place a job into a held state:
$ scontrol hold job_id
We can then release a held job using the release command:
$ scontrol release job_id
scontrol can also provide information on jobs using the show job
command. The information provided by this command is quite extensive
and detailed, so be sure to either clear your terminal window, grep
certain information from the command, or pipe the output to a separate
text file:
# Output to console
$ scontrol show job job_id
# Streaming output to a textfile
$ scontrol show job job_id > outputfile.txt
# Piping output to Grep and find lines containing the word "Time"
$ scontrol show job job_id | grep Time
See also
For a full primer on grep and regular expressions, visit GNU’s page on Grep.
For more information on scontrol, visit the Slurm page on scontrol.