Login

SuperB distributed computing

→ Europe/Rome
Description
VoIP phone conference. Please call one of the numbers in http://server10.infn.it/video/index.php?page=telephone_numbers and insert the code 1993 when asked for.
    • 17:00 → 17:15
      Outlook of SuperB production activities for the TDR in 2009-2010 15m
      Speaker: Mauro Morandin (INFN)
      Slides
    • 17:15 → 17:30
      Exploitation of CNAF and the Italian GRID 15m
      Speaker: Armando Fella (CNAF)
      Slides
    • 17:30 → 18:00
      Short reports on the possibilities of exploiting computing resources available at other sites 30m
      • Caltech 5m
        Speaker: Prof. Frank Porter (Caltech)
      • SLAC 5m
        Speaker: Dr Steffen Luitz (SLAC)
      • McGill 5m
        Speaker: Dr Steven Robertson (IPP/McGill University)
        McGill WLCG grid site currently offline due to migration of resources to CLUMEQ HPC consortium. Will essentially rebuild site from scratch. New site will have lots of resources, but won't be available for at least six months from now. On longer term, can formally request access to resources from this consortium, piggybacking on ATLAS Tier-2 grid site (I've already make initial enquries about this and received a positive response) . http://www.clumeq.mcgill.ca/ Original grid site was within McGill physics department, utilizing dedicated HEP computing cluster. These cpu/disk resources are still available and can potentially be used for SuperB MC production, but only via local production (i.e. not grid-based). Two local clusters available: IBM 1350 cluster based on 2.6GHz (dual-cpu, single-core) Opterons running SLC 4.6 x86_64 (yes, really 64bit OS). Total of 160 cores available, but shared with other HEP projects so can probably access typically 40-50 average for production. Second is older, smaller BABAR-dedicated cluster running 2.0GHz (I think) Opteron 246 processors with SLC4 32-bit. Total of ~16 cores available. Not much, but it can be fully utilized for SuperB. Have successfully installed an run fast sim locally on both clusters (although haven't systematically validated 64-bit vs 32-bit). Might also have access to part of a postdoc for a few months to run local production. I can also inquire regarding other possible Canadian resources. UVic in particular as very strong computing, with ATLAS Tier-2 site and BABAR Tier A and MC production site.
      • Queen Mary 5m
        Speaker: Dr Adrian Bevan (Queen Mary)
        Slides
      • RAL 5m
        Speaker: Fergus Wilson (RAL)
        document
      • LAL and Lyon 5m
        Speaker: Dr NICOLAS ARNAUD (LAL ORSAY CNRS-IN2P3)
    • 18:00 → 18:20
      Discussion 20m