NERSCPowering Scientific Discovery Since 1974

August 30

Attending:  Eric, Katie, Mike, JeffA, Lisa, JeffP, Marjorie, Elizabeth, Larry

Introductions:  Mike is taking over for Andrei in ATLAS, Lisa is taking over for Joanna in icecube.

Cluster Status/Utilization:  Full, roughly half STAR and half ALICE with icecube running consistently, too.  Up to almost 1500 cores now with new ALICE nodes but will shrink back when old nodes are removed.

Recent Outages/Incidents:  /common migration was completed almost 2 weeks ago and it went smoothly.  Also fixed the network switch that had been causing problems in the past.

Upcoming Downtimes:  None.

Grid Issues:  ATLAS transfers are working much better now after things were adjusted at RCF.  Alice is starting to see some issues with the scalability of their grid jobs since they increased the maximum number of jobs allowed. 

I/O benchmarks:  Eric gave a demo of the new I/O benchmarks pages.

Notes from Jay:

- The new compute nodes have been installed and most are in service. We are checking the configuration of about 10 nodes before putting them into service.
- The new storage is mostly installed, and is being tested now before we put it into service (Larry has details if ALICE needs it).

- ATLAS purchases for Fall have been ordered. Waiting for Dell to confirm ETA.

- Dayabay purchases were a little delayed while we worked out support issues for the new storage. We will be ordering those in the next couple of days.