UVA ✎ compute_slurm [Computer Science Support Wiki]

This page is read only. You can view the source, but not change it. Ask your administrator if you think this is wrong.
====== Scheduling a Job using the SLURM job scheduler ======

{{ ::introtoslurm.pdf | Intro to SLURM slides }}

The Computer Science Department uses a "job scheduler" called [[https://en.wikipedia.org/wiki/Slurm_Workload_Manager|SLURM]].  The purpose of a job scheduler is to allocate computational resources (servers) to users who submit "jobs" to a queue. The job scheduler looks at the requirements stated in the job's script and allocates to the job a server (or servers) which matches the requirements specified in the job script. For example, if the job script specifies that a job needs 192GB of memory, the job scheduler will find a server with at least that much memory free.

===== Using SLURM =====
**[[https://slurm.schedmd.com/pdfs/summary.pdf|
Slurm Commands Cheat Sheet]]**

==== Information Gathering ====

SLURM provides a set of tools to use for interacting with the scheduling system.  To view information about compute nodes in the SLURM system, we can use the command ''%%sinfo%%''.

<code>
[pgh5a@portal04 ~]$ sinfo
PARTITION AVAIL  TIMELIMIT  NODES  STATE NODELIST
main*        up   infinite      3  drain falcon[3-5]
main*        up   infinite     27   idle cortado[01-10],falcon[1-2,6-10],lynx[08-12],slurm[1-5]
gpu          up   infinite      4    mix ai[02-03,05],lynx07
gpu          up   infinite      1  alloc ai04
gpu          up   infinite     12   idle ai[01,06],lynx[01-06],ristretto[01-04]
</code>

With ''%%sinfo%%'' we can see a listing of the job queues or "partitions" and a list of nodes associated with these partitions.  A partition is a grouping of nodes, for example our //main// partition is a group of all general purpose nodes, and the //gpu// partition is a group of nodes that each contain GPUs.  Sometimes hosts can be listed in two or more partitions.

To view jobs running on the queue, we can use the command ''%%squeue%%''.  Say we have submitted one job to the main partition, running ''%%squeue%%'' will look like this:

<code>
pgh5a@portal01 ~ $ squeue
             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)
            467039      main    my_job    pgh5a  R      0:06      1 artemis1
</code>

and now that a node has been allocated, that node ''%%artemis1%%'' will show as //alloc// in ''%%sinfo%%''

<code>
pgh5a@portal01 ~ $ sinfo
PARTITION     AVAIL  TIMELIMIT  NODES  STATE NODELIST
main*            up   infinite     37   idle hermes[1-4],artemis[2-7],slurm[1-5],nibbler[1-4],trillian[1-3],granger[1-6],granger[7-8],ai0[1-6]
main*            up   infinite      1  alloc artemis1
qdata            up   infinite      8   idle qdata[1-8]
qdata-preempt    up   infinite      8   idle qdata[1-8]
falcon           up   infinite     10   idle falcon[1-10]
intel            up   infinite     24   idle artemis7,slurm[1-5],granger[1-6],granger[7-8],nibbler[1-4],ai0[1-6]
amd              up   infinite     13   idle hermes[1-4],artemis[1-6],trillian[1-3]
</code>

==== Jobs ====

To use SLURM resources, you must submit your jobs (program/script/etc.) to the SLURM controller.  The controller will then send your job to compute nodes for execution, after which time your results will be returned.

Users can submit SLURM jobs from ''%%portal.cs.virginia.edu%%''.  From a shell, you can submit jobs using the commands [[https://slurm.schedmd.com/srun.html|srun]] or [[https://slurm.schedmd.com/sbatch.html|sbatch]].  Let's look at a very simple example script and ''%%sbatch%%'' command.

Here is our script, all it does is print the hostname of the server running the script.  We must add ''%%SBATCH%%'' options to our script to handle various SLURM options.

<code bash>
#!/bin/bash

#SBATCH --job-name="Slurm Simple Test Job" #Name of the job which appears in squeue
#
#SBATCH --mail-type=ALL
#SBATCH --mail-user=pgh5a@virginia.edu
#
#SBATCH --error="my_job.err"                    # Where to write std err
#SBATCH --output="my_job.output"                # Where to write stdout
#SBATCH --nodelist=slurm1

hostname
</code>

Let's put this in a directory called ''%%slurm-test%%'' in our home directory.  We run the script with ''%%sbatch%%'' and the results will be put in the file we specified with ''%%--output%%''.  If no output file is specified, output will be saved to a file with the same name as the SLURM jobid.

<code>
pgh5a@portal01 ~ $ cd slurm-test/ 
pgh5a@portal01 ~/slurm-test $ chmod +x test.sh 
pgh5a@portal01 ~/slurm-test $ sbatch test.sh 
Submitted batch job 466977
pgh5a@portal01 ~/slurm-test $ ls
my_job.err  my_job.output  test.sh
pgh5a@portal01 ~/slurm-test $ cat my_job.output 
slurm1
</code>

Here is a similar example using ''%%srun%%'' running on multiple nodes:

<code>
pgh5a@portal01 ~ $ srun -w slurm[1-5] -N5 hostname
slurm4
slurm1
slurm2
slurm3
slurm5
</code>

==== Terminating Jobs ====

Please be aware of jobs you start and make sure that they finish executing.  If your job does not converge, it will sit in the queue taking up resources and preventing others from running their jobs.

To cancel a running job, use the ''%%scancel [jobid]%%'' command

<code>
ktm5j@portal01 ~ $ squeue
             JOBID PARTITION     NAME     USER ST       TIME  NODES NODELIST(REASON)
            467039      main    sleep    ktm5j  R       0:06      1 artemis1           <--  Running job
ktm5j@portal01 ~ $ scancel 467039
</code>

==== Partitions ====

Slurm refers to job queues as //partitions//.  These queues can have unique constraints such as compute nodes, max runtime, resource limits, etc.  There is a ''%%main%%'' queue, which will make use of all non-GPU compute nodes.

Partition is indicated by ''%%-p partname%%'' or ''%%--partition partname%%''.

To specify a partition with ''%%sbatch%%'' file:

<code>
#SBATCH --partition=gpu
</code>

Or from the command line with ''%%srun%%''

<code>
-p gpu
</code>

An example running the command ''%%hostname%%'' on the //main// partition, this will run on any node in the partition:

<code>
srun -p main hostname
</code>

==== GPUs ====

Slurm handles GPUs and other non-CPU computing resources using what are called [[https://slurm.schedmd.com/gres.html|GRES]] Resources (Generic Resource).  To use the GPU(s) on a system using Slurm, either using ''%%sbatch%%'' or ''%%srun%%'', you must request the GPUs using the ''%%--gres:x%%'' option.  You must specify the ''%%gres%%'' flag followed by ''%%:%%'' and the quantity of resources

Say we want to use 4 GPUs on a system, we would use the following ''%%sbatch%%'' option:

<code>
#SBATCH --gres=gpu:4
</code>

Or from the command line 

<code>
--gres=gpu:4
</code>

==== Interactive Shell ====

We can use ''%%srun%%'' to spawn an interactive shell on a server controlled by the SLURM job scheduler.  This can be useful for debugging purposes as well as running jobs without using a job script. Creating an interactive session also reserves the node for your exclusive use. 

To spawn a shell we must pass the ''%%--pty%%'' option to ''%%srun%%'' so output is directed to a pseudo-terminal:

<code>
pgh5a@portal ~$ srun -w slurm1 --pty bash -i -l -
pgh5a@slurm1 ~$ hostname
slurm1
pgh5a@slurm1 ~$
</code>

The ''%%-i%%'' argument tells ''%%bash%%'' to run as interactive.  The ''%%-l%%'' arg instructs bash that this is a login shell, this, along with the final ''%%-%%'' are important to reset environment variables that otherwise might cause issues using [[linux_environment_modules|Environment Modules]]

If a node is in a partition other than the default "main" partition (for example, the "gpu" partition), then you must specify the partition in your command, for example:
<code>
pgh5a@portal ~$ srun -w lynx05 -p gpu --pty bash -i -l -
</code>

===== Reservations =====

Reservations for specific resources or nodes can be made by submitting a request to <cshelpdesk@virginia.edu>.  For more information about using reservations, see the main article on [[compute_slurm_reservations|SLURM Reservations]]

===== Note on Modules in Slurm =====

Due to the way sbatch spawns a bash session (non-login session), some init files are not loaded from ''%%/etc/profile.d%%''.  This prevents the initialization of the [[linux_environment_modules|Environment Modules]] system and will prevent you from loading software modules.

To fix this, simply include the following line in your sbatch scripts:

<code bash>
source /etc/profile.d/modules.sh
</code>