NERSCPowering Scientific Discovery Since 1974

September 27

Attending:  Eric Elizabeth, Larry, Lisa, JeffP, JeffA, Craig, Iwona, Mike

Personnel changes:  Iwona to take over as PDSF lead in November with Elizabeth and Larry on the admin side and Eric on the user services side.

Utilization:  Cluster has been running at capacity which is still about 1500 cores.  Mostly STAR and ALICE jobs.

Outages:  None.

Upcoming downtimes:  No date yet but will need a downtime to move the admin nodes and a switch.  Currently waiting to get a new rack.  Also there will be an NGF downtime probably in November.

Procurements:  All new hardware has arrived and we expect to have the new compute nodes in place by 10/17.  New xrootd servers for ALICE will be ready Monday.  Dayabay new disks are awaiting delivery of power strips.

Grid Issues:  JeffP mentioned the scaling issues they run into when the try to run more jobs because of all the little files globus creates.  For now things are running OK but a long term solution will need to be found.

Other Topics:

Craig mentioned that the dayabay data rate is considerable higher than expected and this is putting a strain on their resources.

Mike asked about a way to automatically get ATLAS users certs into NIM but there isn't a way - the users just need to do it themselves.

Elizabeth asked JeffP if they had done any further torque testing and described that we now have reservations working.