Running SAS in parallel using SLUMR job arrays

[dapenDaniel]

Hello. I have a data set with more than 1 million observations. I would like to split this data set and submit them in a job array. However, the computer is using Linux system.

If I submit the single order, The code is 

module load sas
sas /work/sas/code.sas

If I submit the job array, the code is 

module load sas
sas /work/sas/code $SLURM_ARRAY_TASK_ID.sas

However, how can I add a parameter to tell sas that I want sas to read the observation from #1 to #5000, #5001 to #10000 respectively? Thanks.

https://hcc-docs.unl.edu/display/HCCDOC/Submitting+a+Job+Array

[SASKiwi]

A million observations isn't very big, especially for a server. Have you tried a single job and how long did it take to process?

[LinusH]

No idea how SAS would interact with this, assume it needs some investigation to understand, and majenit work.
You could issue parallelization from within SAS using MP CONNECT.