r2 - 16 Feb 2010 - 13:32:58 - KaushikDeYou are here: TWiki >  Admins Web > MinutesDataManageFeb16



Minutes of the US ATLAS Data Management meeting, Feb. 16, 2010
  • Previous meetings and background : IntegrationProgram
  • Coordinates: Tuesdays, Noon Central
    • (605) 715-4900, Access code: 735188; Dial *6 to mute/un-mute.


  • Meeting attendees: Armen, John, Shawn, Bob, Saul, Pedro, Charles, Michael, Wensheng, Patrick
  • Apologies: Wei
  • Guests:

Topics for this week

  • SCRATCHDISK filled up at BNL - Armen
    • DQ2 uses SCRATCHDISK as buffer for when transferring files to Tier 2's.
    • Armen checked - very few user files, mostly old files (which should have been cleaned up after 30 days), and temporary files.
    • Central deletion needs to clean up - started by Stephane already today, after prompting by Armen? Armen will follow up.
    • Armen also found a lot of dark data.
    • Charles - will add new options in ccc.py to specify date (older than) and spacetoken name.
  • GROUPDISK filled up at BNL - Armen
    • From Armen - we had more than 8TB available space in GROUPDISK yesterday, until it was an attempt to stage 14TB dataset from BNL-OSG2_DATATAPE to BNL-OSG2_PERF-MUON!
    • From Kaushik - see http://panda.cern.ch:25980/server/pandamon/query?mode=ddm_req&req_id=4234&action=Info, approved by domizia.orestano @ cern.ch
    • From Kaushik - new policy requested by Stephane, that cloud (i.e. KD) no longer needs to approve such large requests?
    • Kaushik - DaTRI? should have checked. Will follow up.
  • MCDISK space Usage - Armen
    • Follow up from last week
      • Add more space to MCDISK. Done.
      • Breakdown of data on MCDISK by project name. Next week - Armen will make table.
      • Check if ESD's generated in the US are being kept on tape - Kaushik.
      • Can we reduce space usage at MWT2? After that NET2? WT2? AGLT2? SWT2?
  • PRODDISK cleaning and consistency - Charles
    • Is new version verified? Yes and Yes. 1.15 should be used by everyone.
  • Hot issues
    • From last week - LOCALGROUPDISK
      • Who should be allowed - next week?
      • Wensheng - 3.5 TB used on LOCALGROUPDISK at BNL, out of which 2.1 TB are actually official production files. Metadata says custodial (random check of one dataset)! Check all dataset locations.
        • 77 datasets: 1 MC, 76 data*. MC dataset has replica in FZK. Other 76 data* datasets are on BNL DATATAPE.
        • Archive bit is custodial for all datasets. Wensheng will follow up.
  • AOB
    • Armen - emails were sent to users last week about USERDISK cleanup. Actual deletion will start next week.

-- KaushikDe - 16 Feb 2010

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