We are currently in the process of tuning how cluster usage impacts users’ FairShare priority factor. Historically, the FairShare priority factor provided a boost to job priority for users with low recent cluster utilization, but droped off sharply after a small amount of cluster resources were consumed. This tuning aims to increase the amount of cluster resources users’ may consume before their FairShare priority drops off.

You may notice changes to job wait times during this tuning. If you have any questions about this change, please contact us. For more information about FairShare and other factors that affect job priority, please refer to our documentation.