[slurm-users] slurm array with non-numeric index values

[Willy Markuske]

Michael answered your question but since I completed my response just as
it came through I guess I will still send it. The only thing I would
reiterate over Michael's response is you can set the slurm array id's as
desired so you don't necessarily have to do the subtract 1 add 1 to
determine array limits.

The situation you are describing is an excellent use for array jobs. You
can create a list of the non-numerical values you want to use and then
call the specific one you want using the built in jobid environment
vairables. The list can be created as a bash list in your submission
script directly or an input file based on what is easiest to make.

Example: Assume I have 4 different input values I want to run for the
same script. I would create a submission script as such:

#!/bin/bash
#SBATCH --array=0-3
inputs=('in1' 'in2' 'in3' 'in4')
script.py ${inputs[$SLURM_ARRAY_TASK_ID]}

This will submit 4 separate jobs running different inputs for your
python script.

If you have 500 individual jobs that you want to run putting all you
inputs in your submission script is probably a lot of unnecessary work
and would make changing your inputs difficult. In that case you can make
a file where each input is on it's own line and create a submission
script similar to:

#!/bin/bash
#SBATCH --array=1-4
file='inputfile'
input=$(sed -n -e ${SLURM_ARRAY_TASK_ID}p $file
script.py $input

Be aware that bash lists start indexing at 0 so your array would likely
start at 0 and sed line numbers will start at 1 so you want to start at
1. You can also make the array index to be anything you want so if you
know you only want to run lines 100-200 you can set array=100-200.

Regards,

Willy Markuske

HPC Systems Engineer

	

Research Data Services

P: (858) 246-5593

On 7/15/20 1:45 PM, Renfro, Michael wrote:
> If the 500 parameters happened to be filenames, you could do adapt like (appropriated from somewhere else, but I can’t find the reference quickly:
>
> =====
>
> #!/bin/bash
>  # get count of files in this directory
> NUMFILES=$(ls -1 *.inp | wc -l)
> # subtract 1 as we have to use zero-based indexing (first element is 0)
> ZBNUMFILES=$(($NUMFILES - 1))
> # submit array of jobs to SLURM
> if [ $ZBNUMFILES -ge 0 ]; then
>   sbatch --array=0-$ZBNUMFILES array_job.sh
> else
>   echo "No jobs to submit, since no input files in this directory.”
> fi
>
> =====
>
> with:
>
> =====
>
> #!/bin/bash
> #SBATCH --nodes=1  --ntasks-per-node=1 --cpus-per-task=1
> #SBATCH --time=00:01:00
> #SBATCH --job-name array_demo_2
>  
> echo "All jobs in this array have:"
> echo "- SLURM_ARRAY_JOB_ID=${SLURM_ARRAY_JOB_ID}"
> echo "- SLURM_ARRAY_TASK_COUNT=${SLURM_ARRAY_TASK_COUNT}"
> echo "- SLURM_ARRAY_TASK_MIN=${SLURM_ARRAY_TASK_MIN}"
> echo "- SLURM_ARRAY_TASK_MAX=${SLURM_ARRAY_TASK_MAX}"
>  
> echo "This job in the array has:"
> echo "- SLURM_JOB_ID=${SLURM_JOB_ID}"
> echo "- SLURM_ARRAY_TASK_ID=${SLURM_ARRAY_TASK_ID}"
>
> # grab our filename from a directory listing
> FILES=($(ls -1 *.inp))
> FILENAME=${FILES[$SLURM_ARRAY_TASK_ID]}
> echo "My input file is ${FILENAME}”
>
> # make new directory, change into it, and run
> mkdir ${FILENAME}_out
> cd ${FILENAME}_out
> echo "First 10 lines of ../${FILENAME} are:" > ${FILENAME}_results.out
> head ../${FILENAME} >> ${FILENAME}_results.out
>
> =====
>
> If the 500 parameters were lines in a file, the same logic would apply:
>
> - subtract 1 from the number of lines in the file to determine the array limit
> - add 1 to ${SLURM_ARRAY_TASK_ID} to get a line number for a specific parameter
> - something like "sed -n ‘${TASK_ID_PLUS_ONE}p’ filename” to retrieve that parameter
> - run the Python script with that value
>
>> On Jul 15, 2020, at 3:13 PM, c b <breedthoughts.www at gmail.com> wrote:
>>
>> External Email Warning
>> This email originated from outside the university. Please use caution when opening attachments, clicking links, or responding to requests.
>> I'm trying to run an embarrassingly parallel experiment, with 500+ tasks that all differ in one parameter.  e.g.:
>>
>> job 1 - script.py foo
>> job 2 - script.py bar
>> job 3 - script.py baz
>> and so on.
>>
>> This seems like a case where having a slurm array hold all of these jobs would help, so I could just submit one job to my cluster instead of 500 individual jobs.  It seems like sarray is only set up for varying an integer index parameter.  How would i do this for non-numeric values (say, if the parameter I'm varying is a string in a given list) ?
>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20200715/181a78b7/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: SDSClogo-plusname-red.jpg
Type: image/jpeg
Size: 9464 bytes
Desc: not available
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20200715/181a78b7/attachment-0001.jpg>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 488 bytes
Desc: OpenPGP digital signature
URL: <http://lists.schedmd.com/pipermail/slurm-users/attachments/20200715/181a78b7/attachment-0001.sig>