LCG Experiment Coordination Meeting

Europe/Zurich
28-R-015

28-R-015

Description
Coordination meeting between LCG service and LHC experiments, focussing on issues for the coming weeks / months.

Subscribe to lcg-experiment-coordination@cern.ch to be kept informed...

To join the call, do one of the following:
a. Dial +41227676000 (Main) and enter access code 0166222, or
b. To have the system call you, click here: https://audioconf.cern.ch/call/0166222

(Leader code is 0166111)

    • 14:30 14:40
      FTS 2.0 pilot service and schedule / actions for move to production 10m
      At this meeting we need to conclude on experiment-testing of FTS 2.0, if we are to stick to the agreed timetable of:

      • circa 1 month testing at CERN, including 1-2 patch updates;
      • roll-out at all Tier1s to allow for the services to be fully debugged and settled in prior to dress rehearsal activities

      This would translate to:

      • June 11th 2007: experiments give the green light to FTS 2.0 deployment in production at CERN;
      • June 18th 2007: upgrade of CERN services (TBC)
      • July 9th 2007: green-light to Tier1 sites to start deployment of FTS 2.0 services at their sites;
      • July/August: settling in and debugging at sites as required;
      • Post-CHEP 2007: full production load.

      The current status of experiment testing / validation (AFAIK) is:

      • ALICE: ok given at June GDB;
      • ATLAS: ok;
      • CMS: ok;
      • LHCb: ok.
    • 14:40 14:45
      FTS service monitoring requirements 5m
      From CMS "top 5" issues presented to MB in Prague:
      • transfers and queues monitor
      • backup servers
      • heartbeat and similar ”service-is-up" monitoring

      Can we come to a common requirement / proposal?

    • 14:45 14:50
      Status of WLCG Q1 Tx-Ty Transfer Milestones - aka Multi-VO tests 5m

      Proposal from June GDB is to attempt Tier0-Tier1 export for one week at the following daily averages:

      • ATLAS - 650MB/s
      • CMS - 200MB/s
      • ALICE & LHCb - 30MB/s each

      Any fluctuations in export rate need to be explained.

      1. Demonstrate Tier0-Tier1 data export at 65% of full nominal rates per site using experiment-driven transfers

        Mixture of disk / tape endpoints as defined by experiment computing models, i.e. ~30% tape for ATLAS; transfers driven by experiments

        Period of at least one week; daily VO-averages may vary (~normal)

      2. Demonstrate Tier0-Tier1 data export at 50% of full nominal rates (as above) in conjunction with T1-T1 / T1-T2 transfers

        Inter-Tier transfer targets taken from ATLAS DDM tests / CSA06 targets

      3. Demonstrate Tier0-Tier1 data export at 35% of full nominal rates (as above) in conjunction with T1-T1 / T1-T2 transfers and Grid production at Tier1s

        Each file transferred is read at least once by a Grid job

        Some explicit targets for WMS at each Tier1 need to be derived from above

    • 14:50 15:05
      SRM v2.2 site / experiment test schedule and July pre-GDB meeing 15m
    • 15:05 15:20
      SLC4 SL3-compatibility Worker Nodes 15m
      So far access to the SLC4 SL3-compatibility Worker Nodes has only been available through the pre-production service. CERN plans to install the SLC4 32-bit native build gLite worker node middleware now it has been released and put them under a production CE. This will take several weeks and we would like to know if/when the experiments are ready for this. Jobs which can only run on SL3 will have to request this in their JDL. See: http://goc.grid.sinica.edu.tw/gocwiki/How_to_publish_my_machine_architecture http://goc.grid.sinica.edu.tw/gocwiki/How_to_publish_the_OS_name
      Speaker: Ulrich Schwickerath (CERN)
    • 15:20 15:25
      Up-coming service interventions 5m

      Possible slots prior to start of Dress Rehearsals and Pilot LHC run:

      • Slot in June, e.g. June 11 (following weeks are CMS / ATLAS weeks respectively).
      • Week of CHEP 2007?? (1st week of September)

      Foreseen interventions:

      • Move of CASTOR DBs to better h/w (to be done by VO - service interruption involved). Now fully done - item to be removed.
      • Deployment of new CASTOR version, including new LSF plug-in (ATLAS, CMS, PUBLIC, LHCb, ...)

        Proposed schedule:

        • PUBLIC: June 12
        • ATLAS : June 13
        • LHCB : June 18
        • CMS : TBD
        • ALICE : TBD

      • Oracle Critical Patch updates
        • 17 April 2007 (skipped)
        • 17 July 2007
        • 16 October 2007
        • 15 January 2008
    • 15:25 15:30
      AOB 5m