• Running job or CPU intensive process on scail.stanford.edu (headnode) itself is strictly prohitbited, we will kill such processes without warning.

  • Running job or processes outside the queuing system is strongly discouraged, sometime we kill such processes without warning if they are causing trouble to other users.

  • If you need to run your job outside the queuing system and/or need to reserve a batch of machines, please contactaction@ai.stanford.edu, we will review the situation and properly assign you a batch of machines for exlusive use.

  • When running job that requires access to AFS for more than 24 hours, please run 'reauth' before doing a qsub, your AFS token will be renewed after 24 hours.

  • In the current torque setup, a message that says "PBS: chdir to /afs/http://cs.stanford.edu/u/csid failed: Permission denied (running user prologue in current directory)" will appear after you successfully qsub, please ignore this message as it should have no impact to your job.