I am facing an issue with the time taken by DevCloud to run a job.
The time taken to run a job is easily 7 times more than it used to take a week back for unchanged code.
Also, there was no change in the run time when the nodes were allocated or not.
This might be because of the node allocation issue or something else related to node network.
It would be very helpful if you can check this and revert back as soon as possible.
For your information,
The command-line which I used to allocate node is
qsub -I -l nodes=4:ppn=2
Does the issue still persist? If yes, please try submitting the job at different times and check whether the issue still exists. Please let us know the same
Yes, the issue still persists and keeps existing for a run at different times.
Please look at the issue again and revert me back.
Thank you for your effort, but as of now, I don't find time gain while running the job with unchanged code.
It is taking the same time as it used to take earlier i.e 7 times more than the usual case.
I will do a check again and will confirm it by tomorrow.
We apologize for the delay. According to the response we got from the DevCloud team, DevCloud does not make any assurance for how long it takes for a resource to be allocated to a job. A job can be configured to be triggered automatically after another job completes successfully. So, It completely depends on the nodes available at that particular time so that they can get allocated to a new job.