Why is my job still waiting in the queue?

The following factors may contribute to longer wait time, or jobs not starting (depending on your account’s access level):

• Larger memory request, e.g. (h_data) or (h_data)*(-pe shared)
• Longer run time (h_rt)
• Specific CPU model (arch)
• Many CPUs (-pe dc*)
• You are already running on some numbers of CPUs or nodes
• For high priority jobs (-l highp):
  • Your group members are already running on your purchased nodes; there are not enough left for your job to start
  • Your request exceeds what your group nodes have available
    

Article on the Hoffman2 Website

Current Hoffman2 Cluster Utilization