1 / 11

Backbone Analysis Grid

Backbone Analysis Grid. A Skeleton for LHCb? LHCb Grid Meeting Bologna, 14th June 2001 Glenn Patrick (RAL) http://hepwww.rl.ac.uk/lhcb/physics/bologna140601.ppt. MPG Monte-Carlo Production Grid. Monte-Carlo Production is readily mapped onto a Grid Architecture because:

field
Download Presentation

Backbone Analysis Grid

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Backbone Analysis Grid • A Skeleton for LHCb? • LHCb Grid Meeting • Bologna, 14th June 2001 • Glenn Patrick (RAL) • http://hepwww.rl.ac.uk/lhcb/physics/bologna140601.ppt Glenn Patrick

  2. MPGMonte-Carlo Production Grid • Monte-Carlo Production is readily mapped onto a Grid • Architecture because: • It is a well defined problem using the same executable. • Already requires distributed resources (mainly cpu) in large centres (eg. Lyon, RAL, Liverpool ...). • Few people involved. With Eric’s help, we all learnt a lot from the mini project between CERN, RAL, Liverpool et al. Particularly:What worked, what didn’t work, what was hype,the Grid products we need to wait for. Glenn Patrick

  3. Exists Planned Initial LHCb MPG “Testbed” First lesson! Not much of this worked at the same time for all people. CERN pcrd25.cern.ch lxplus009.cern.ch RAL CSF 120 Linux cpu IBM 3494 tape robot Institutes LIVERPOOL MAP 300 Linux cpu RAL (PPD) Bristol RAL DataGrid Testbed Imperial College GLASGOW/ EDINBURGH “Proto-Tier 2” Oxford Cambridge Glenn Patrick

  4. BAG Backbone Analysis Grid Analysis is much more inventive, chaotic and apparently controversial. It will involve far more people in a wide range of institutes. • We need to get a feel for the issues. • We need to be realistic. • Gaudi will need a skeleton testbed to develop any Grid services. • We need a basic Grid structure to try things out (and be able to get them wrong as well as right!). Glenn Patrick

  5. “I typed my Grid ID into the wrong box and accidentally booked us into 90543678910 hotel rooms!” Glenn Patrick

  6. Gaudi meets BAG? FRANCE? Gaudi Services Application Manager Job Options Service Detector Description EventData Service Histogram Service Message Service Particle Property Service GaudiLab Service ITALY? Meta Data Data ?? CERN UK? ?? Logical DataStores Event Detector Histogram Ntuple Standard Interfaces & Protocols Glenn Patrick

  7. DataGrid Testbed Sites BAG is a small subset of this (>40) Dubna Lund Moscow Estec KNMI RAL Berlin IPSL Prague Paris Brno CERN Lyon Santander Milano Grenoble PD-LNL Torino Madrid Marseille BO-CNAF HEP sites Pisa Lisboa Barcelona ESRIN ESA sites Roma Valencia Catania

  8. Lab m Uni x Uni a UK USA FermiLab Lab a France Tier 1 Uni n CERN Tier2 Physics Department Italy Desktop Lab b NL Lab c  Uni y Uni b   LHC Computing Model USA Brookhaven ………. Germany

  9. Focus for other Centres Possible UK Example UK BAG Node (eg. RAL) Tier 2 (eg. Liverpool) Tier 2 (eg. Edin./Glasgow) Tier 3 Glenn Patrick

  10. Requirements and Aims • Select a v.small number of sites (to begin with). • Grid-enabled with standard Globus kit, GSI-ftp, etc. • LHCb software installed in ~ “standard” environment. • Replicated datasets installed at each site (small, disk resident to start with). • Authorisation/authentication for “outsiders”. Provides “mini-grid” to explore distributed analysis.  New LHCb Grid staff (eg. Oxford WP8 post) can help to develop and exploit this infrastructure. •  See what works before worrying about the larger-scale! Glenn Patrick

  11. Conclusions • BAG could provide: • 1. Better understanding of how a distributed Data Model would really work for analysis. Test different models. • 2. A (skeleton) testbed for developers. • 3. Experience of running Gaudi in a distributed LHCb environment. Performance/limitations? • 4. Better understanding of technical issues. • 5. Pull together some of the wider Grid work for benefit of LHCb. • 6. Useful structure for funding/manpower requests? Glenn Patrick

More Related