r2 - 22 Aug 2011 - 13:07:02 - RichardMountYou are here: TWiki >  AtlasSoftware Web > Minutes27May2011

Minutes27May2011 RAC Minutes, May 27, 2011


Members (*=present, #=apologies)

*Richard Mount, *Jim Cochran (Analysis Support Group Chair), Alexei Klimentov (ATLAS ADC), Ian Hinchliffe (Physics Advisor), Rik Yoshida (Analysis Support Group Deputy), *Michael Ernst (U.S. Facilities Manager), *Rob Gardner (Integration Coordinator), #Kaushik De, (U.S. Operations Manager), *Armen Vartapetian (U.S. Operations Deputy)

Ex-Officio: Torre Wenaus, Stephane Willocq, Mike Tuts, Howard Gordon

Approval or Correction of Minutes

The minutes of May 6 were approved

Operations Report (Armen)

Production

MC10 finished at the end of last week after some uneven running. A new technical issue was seen - jobs in the queue but not running. Seeing this needs database access. MC11 tuning is still being discussed. Final MC11 validation was supposed to take place this week but its outcome is unknown.

Analysis

Analysis continues at the relatively low level seen in recent weeks.

Data Management

Generally OK, but the central deletion rate at BNL continues to be a concern. Deletion was running at 6-7 Hz, but has now fallen back to 1-2 Hz. This mystery is under investigation. BNL can delete locally.

A new round of USERDISK deletion has started. This needs to run at the full rate to complete in 2-3 weeks.

A new scheme for storage reporting will include unallocated and unpowered storage at sites.

Michael: A new version of site services allowing US T2s to communicate with any T1 was deployed last week. The job-distribution has not changed - the majority of analysis jobs go to T1s.

Request for Additional SUSY Production

The request was relayed by Kaushik and is attached to the Indico agenda.

The RAC strongly endorsed this request.

Issues arising from the CREM approval of TRIG-DAQ space at MWT2

Richard described the CREM discussions. CREM members were concerned about potential rapid growth of GROUPDISK endpoints - probably because the logic behind the initial assignments was only vaguely remembered and it was not clear how to say no to new requests. There would be no objection to moving existing endpoints to different sites within the US cloud. It was accepted that this would take too long for the TRIG-DAQ/MWT2 need, but the US was asked to examine whether all other TRIG-DAQ spaces in the US were needed and used.

Formally, the CREM required GROUPDISK endpoint requests to come from the group concerned. In this case, Chris Bee had already given his support so the CREM agenda was changed to make the request come from TRIG-DAQ. In addition, it was advised and accepted to drop the text about exploiting the MWT2 endpoint from a co-located T3.

In discussions, Michael proposed merging GROUPDISK and DATADISK, and creating a much more dynamic system. Michael believed that most of the tools were in place to manage quotas for the groups. The best way to overcome any remaining problems would be to do the merge and address problems as they arose. Flexibility to place specific data at specific sites should remain where needed. An ATLAS-wide approach was needed - the US should not go it alone.

AOB

None

Action Items

  1. 5/6/2011: Richard: Draft a web page explaining how users should exploit the various SpaceTokens - Ongoing.
  2. 5/6/2011: Kaushik/Richard: Bring the request for a TRIG-DAQ GroupDisk at MWT2 to the CREM - Done.
  3. 3/25/2011: Richard: Email US ATLAS about the availability of production and analysis resources; ask Jim/Ric why there is so little analysis activity - Ongoing.
  4. 9/24/2020: Kaushik: Organize the nature and timing of the effort to create an automated mechanism to give US-requested additional production priority access to US non-pledged resources. On hold - manual system works just fine for now

About This Site

Please note that this site is a content mirror of the BNL US ATLAS TWiki. To edit the content of this page, click the Edit this page button at the top of the page and log in with your US ATLAS computing account name and password.


Attachments

 
Powered by TWiki
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback