May 21 – 24, 2018
Fluno Center on the University of Wisconsin-Madison Campus
America/Chicago timezone

Prioritizing vanilla and grid jobs from local users on a Tier-3 condor cluster

May 22, 2018, 2:10 PM
20m
Howard Auditorium (Fluno Center on the University of Wisconsin-Madison Campus)

Howard Auditorium

Fluno Center on the University of Wisconsin-Madison Campus

601 University Avenue, Madison, WI 53715-1035

Speaker

Kenyi Hurtado Anampa (University of Notre Dame)

Availability of the Speaker<br>Let us know if there are times you CANNOT present,<br>prehaps because you need to leave for the airport early, etc.

My only constraint would be on Thursday afternoon, as I might be leaving in the morning. So, I would avoid presenting that day if possible.

Summary (2-4 sentences)<br>Just a few informal sentences describing what you want to present.<br>No need to spend a lot of time on this! You can change it later.

The Notre Dame CMS Tier-3 condor cluster is used for ND users to submit analysis jobs locally or through the grid (from e.g CERN). We also share the cluster with OSG, CMS and other VOs opportunistically through an HTCondor-CE. This work was motivated by the fact we want any ND CMS user submitting in any way (vanilla or grid universe) to have full priority on the resources at all times, quickly preempting or draining machines with any non-ND grid user if necessary. For draining machines, we also do some segregation in the pool ( so machines running ND jobs, prefer other ND jobs and grid jobs are also grouped together in machines with no ND jobs when possible).

Primary author

Kenyi Hurtado Anampa (University of Notre Dame)

Presentation materials