1 / 10

LHC Computing MoU(s)

LHC Computing MoU(s). April 2004 (Computing) RRB discussion. Status of proposed Computing MoU for LCG (LHC!?) Suggestions and worries (INFN, CERN and Experiments roles) Conclusions and request for actions (by CSN1 and INFN) Current view by Italian components of

Download Presentation

LHC Computing MoU(s)

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. LHC Computing MoU(s) April 2004 (Computing) RRB discussion • Status of proposed Computing MoU for LCG (LHC!?) • Suggestions and worries (INFN, CERN and Experiments roles) • Conclusions and request for actions (by CSN1 and INFN) • Current view by Italian components of • ATLAS (L. Perini), CMS (P. Capiluppi) and LHCb (U. Marconi) • (+ Alice (M. Masera))

  2. MoU(s) Environment • Many actors and players • Experiments (ATLAS, CMS, LHCb, ALICE) • Funding Agencies (~30 worldwide) • CERN • LCG (LHC Computing Grid Project) • Grid (national and international) Projects • Computing TDRs (or Computing Model Definition) • Affordable Computing Model(s) specifications • Amount of needed resources and level of services (costs?) • Due ~mid 2005 for Experiments and LCG • Construction of part of the LHC Detectors • Years 2005/6 ~ 2010/11 • Costs ranging from: [early estimates!] • Les: 68.8 MCHF/(four experiments) at CERN for 06-08, 34.5 MCHF for 09-10 • CMS Italy: 26.7 MCHF/INFN 06-08 for 1 Tier1 & 6 Tier2s, +3.8 MCHF/INFN for 4 Tier3s

  3. LHC Computing MoU Task force • CERN: • Jos Engelen, Hans Hoffmann, David Jacobs • Experiments: • Torsten Akesson (ATLAS), Nick Brook (LHCb), Yves Schutz (ALICE), Ian Willers (CMS) • LCG: • Ian Bird, Chris Eck, (Les Robertson) • Countries: • Manuel Delfino (rest of Europe), Irwin Gaines (USA), Neil Geddes (UK), Marcel Kunze (D), Federico Ruggieri (I), Guy Wormser (F) Schedule • 19 February: First meeting • 14 April: Next meeting • 27 April: Spring RRB: first report on taskforce work • Autumn RRB: taskforce presents structure of LHC Computing MoU • Spring 2005: RRB: final document • June 2005: signatures requested • MoU becomes active already in 2005

  4. Current Draft ~30 pages, (too?) many annexes Memorandum of Understanding for Collaboration in the Deployment and Exploitation of the LHC Computing Grid (LCG) between The EUROPEAN ORGANISATION FOR NUCLEAR RESEARCH, hereinafter referred to as CERN, Geneva, as the Host Laboratory and provider of the Tier0 Centre, provider of the CERN Tier1 Centre and coordinator of the LCG project on the one hand and an Institution/Funding Agency contributing to a Tier1 Centre of the LHC Computing Grid on the other hand.

  5. LCG Collaboration ! (Phase II) • Article 1 : Parties to this MoU • 1.1 The Parties shall be all the Tier1 Centres listed in Annex 1 and their Funding Agencies, and CERN as a) the Host Laboratory and provider of the Tier0 Centre, b) provider of a Tier1 Centre and c) coordinator of the LCG project. Annex 1 shows for each centre the person responsible towards the Collaboration…. Annex 2 lists the Funding Agencies and their duly authorized representatives to the C-RRB. • Article 2 : Purpose of this MoU • 2.1 This MoU addresses the deployment and exploitation phase of the LCG project (Phase 2), and in particular the Tier0 Centre and the Tier1 Centres that provide the mechanism for delivering computing resources in a common or shared way to the LHC experiments. … It sets out organisational, managerial and financial guidelines to be followed by the Collaboration.

  6. LCG Phase II Computing Estimates • Current Tier 1 Centres • a. CNAF – all experiments • b. PIC – ATLAS, CMS, LHCb • c. Lyon – all experiments • d. GridKA – all experiments • e. RAL – all experiments • f. BNL – ATLAS • g. FNAL – CMS • h. Tokyo – ATLAS (to be confirmed) • I. CERN - all experiments • Experiment representatives to define classes of work to be done at Tier 1 centres and at Tier 2 centres • Experiment representatives to produce capacity estimates for Tier 1 and Tier 2 centres. • Estimates based on amount of resources to be committed • Order of ~100 Tier2s for the four LHC Experiments worldwide

  7. It’s a Construction-like MoU ?(2005-11) • Parties and Signatories: (Customer(s) and Supplier(s)?) • The "contract" specifies the services to be provided, associated costs and the technical and financial responsibilities. • The signatory representing the customer(s), i.e. the Experiment(s), are: • CERN as host laboratory • AND the Funding Agencies. • Signatories representing the suppliers may in some cases be: • the suppliers themselves (e.g. CERN); • or a suppliers coordinator (e.g. CERN representing the LCG); • or the Funding Agency for a supplier of resources Tier1s and Tier2s (including person-power). • The (technical) suppliers are a number of parties, including: • CERN as provider of Tier0, Tier1 and other on-site services; • CERN/LCG project - related to above but also including middleware and applications software, etc.; • Regional Centres (Tier1, Tier2, etc.) via the Funding Agencies; • Other providers: software/middleware developers, non-LHC RC's... Tier1s and Tier2s Funding Agencies providers’ must sign individually!

  8. How many MoUs • A number of choices: • one MoU with Annexes for the 4 Experiments, or • four MoU's with very similar or identical text and differing numbers, or • four small (M&O annexes) MoUs + one biggest MoU for common resources. • Resource Management (who is going to “control”?) • The Experiments and their overseeing bodies (LHCC, etc.) should make decisions about physics optimisation which may involve difficult tradeoffs between computing, the subdetectors, triggers, and even LHC machine operations. • All resources (including Computing) should always be optimised to maximise the Physics output of the LHC. The correct bodies to do this are the Experiments themselves under the oversight of the LHCC and the RRB, following the model of the detector MoU's. A single MoU to be signed by INFN?!

  9. Considerations • INFN • LCG is a provider of infrastructure (technical) coordination for resources • Experiments • MoUs are the means to guarantee the level of competition for Physics discovery from the data of the Detectors • Funding Agencies • Detector funding and responsibilities must rely on committees and scientific decisions • CERN • Host Laboratory for Tier0, one Tier1 and coordinating (LCG) infrastructure responsibility Need for a “clear” statement at the coming RRB by INFN!

  10. Issues and Proposal Experiments need (want) a Computing MoU The MoU must be as all others MoUs: a commitment to provide a “part” of the detector The MoU must be signed by all the committing bodies The scope of the MoU is to build the Computing System for an Experiment The limits (of commitment: money, personnel, infrastructure, services) should be stated in the MoU for each Experiement, and revised by the due Bodies (RRB) regularly All the Funding Agencies must sign for “down” to the Tier2 commitment (~30% of effort) Technical and managerial organization for the Computing System is an annex of the MoU (LCG is the way to do that) • CERN signs as the provider of: Tier0, Tier1 and LCG coordination Project • Funding Agencies sign as provider of: Tier1s, Tier2s and personnel core software • All sign on behalf of the Experiments

More Related