NERSCPowering Scientific Discovery Since 1974

August 2

Attending:  Eric, Jay, Craig, JeffA, JeffP, Marjorie

Utilization:  Cluster mostly full, some free cycles on the weekends.  Mostly STAR jobs but quite a lot of groups have been running some.

Recent Outages:  Nothing major in the past two weeks.

Upcoming Downtimes:  We will schedule something soon probably for the week of 8/15.  Probably an eight hour downtime to complete the /common migration as well as redo some network cabling related to the problems a few weeks ago (borrowed cable from network group).

Procurements/New Hardware:  Requisition will go out today or tomorrow at the latest.  Mostly we're getting nodes but also some networking equipment and miscellaneous as well as a second dtn.  ATLAS, STAR, ALICE and dayabay were sent a spreadsheet detailing what the new shares will be.  They will be put in place after old hardware is retired in the next month or so.    Jay is in the process of setting up the new ALICE nodes and xrootd servers.  This will add about 300 cores.  JeffP requested that ALICE get additional shares when they come online as they are under pressure to provide resources and Jay agreed.

Grid Issues:  There was an incident where a disk filled up on the PDSF dtn that was addressed.  We are getting better at working with the ATLAS grid - Jay reported that he blacklisted our site before our last downtime and Eric and Jay reported that a W channel (for west coast) was created at BNL so that ATLAS transfers wouldn't be competing with other groups (they have the W channel all to themselves for now).

Other Topics:  Craig brought up the benchmarking jobs that dayabay provided.  Eric reported that he is using them but they aren't being displayed on the web yet.  ALICE and STAR jobs in place as well and Marjorie to provide an ATLAS job or two.

JeffP inquired about the status of the torque cluster - it's up now but reconfiguring.  It will be ready later today with the moab part in place.