r3 - 16 Apr 2010 - 18:58:09 - RichardMountYou are here: TWiki >  AtlasSoftware Web > Minutes9Apr2010

Minutes9Apr2010 - RAC core team minutes, April 9, 2010

Core Team Participation

Richard Mount, Kevin Black (for Jim Cochran), Michael Ernst (not available), Kaushik De (Operations)

Other Participants

Torre Wenaus

Report from the CREM Meeting

The documents and emails discussed at the CREM meeting are available on the agenda page: http://indico.cern.ch/conferenceDisplay.py?confId=90014

Richard had attended the CREM meeting (Kaushik was teaching).

The principal CREM topic was the discussion of a draft data deduction and replica reduction policy written by Alexei. The spirit of the draft was accepted, with some detailed requests for clarified language (see the email from Stephane Jezequel). There was some discussion of whether "custodial" copies must always reside on tape without coming to a clear conclusion.

The CREM then discussed an issue raised by Kaushik by email - Many analysis jobs requesting Data10 and MC10 minimum bias ESDs were timing out at BNL because the only ESD copy in the US was at BNL. Kaushik noted that he could fix this my manually replicating ESDs but requested that this be done automatically for the next few weeks. This proposal was accepted for Data (but MC would still only be replicated on demand). It was noted that one Tier-2 replica of the Data10 ESDs should have been available in the US, but a distribution problem (to AGLT2) seemed to be preventing the replication.

Kaushik reacted favorably to the CREM decision. Manually replicating the simulation on demand was easy. Automating additional data ESD distribution was what was needed. The problem with distribution to AGLT2 was under investigation.

Kaushik agreed to try to find out who was running on the ESDs at BNL.

Progress on "Additional Production"

Kevin proposed the creation of a Twiki guiding US physicists through the process of requesting additional production. The process should require a concise physics justification, precise details of what and how much should be simulated, and a procedure for getting approval from the appropriate physics group and getting the request into the task list of central production. It would also need to indicate the level of effort that should be found within the physics group to oversee the production.

Kevin agreed to make a first version of the Twiki. He also agreed that pushing though some simulations required by the Exotics group would provide excellent test cases to exercise and help define the procedures. These would involve both more events at points in parameter space that had already been simulated and new points in parameter space.

The issue of disk space for the additional production was discussed. Since such production would go only to GroupDisk, it would be necessary to increase the disk space allocated to GroupDisk.



Action Items

  1. 4/9/2010: Kaushik, try to find out who is generating the large minimum bias load seen by BNL..
  2. 4/9/2010: Kevin, Create first version of a Twiki guiding US physicists on requesting Additional Production
  3. 3/26/2010: Richard, Further publicize "US Regional Production: Summary of ATLAS Policy and US Practice" (revisit this item after the ATLAS SW and Computing week in April).
  4. 3/26/2010: All but especially core team, Find time during the ADC Workshop next week to identify a point-of-contact for Valid US Regional Production (revisit this item after the ATLAS SW and Computing week in April).
  5. 3/26/2010: Kaushik, Make formal request to the DQ2 team for a tool to display a record of what has been deleted when.
  6. 2/12/2010: Richard Non-urgent simulation tasks - consult with physics groups on the best way to establish two-way communication with the RAC to meet this and other needs (Largely complete - will propose to drop at next full RAC meeting)

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.


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