I am wondering if there are ways to improving the job start time (the time between submit and actual startup). My plan is to use condor to run queue-processors, which are submitted by a manager that makes sure we keep up with the queue. The manager alsoruns in the cluster.
One [hackey] way to do this is to set up your schedd as a dedicated scheduler, as per the manual, and submit your jobs to the "parallel" universe with a machine_count of 1. After the initial claiming the dedicated scheduler holds on to claims for a configureable amount of time (by default, five minutes), so subsequent submits run very fast.
-Greg