les robertson cernit 1 - PowerPoint PPT Presentation

Loading...

PPT – les robertson cernit 1 PowerPoint presentation | free to download - id: 26dabb-YmZhM



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

les robertson cernit 1

Description:

computing fabric management Wolfgang von R den. Chief technical officer Dave Foster ... Goal: Arrive at high level work-plan. Define scope and priorities ... – PowerPoint PPT presentation

Number of Views:48
Avg rating:3.0/5.0
Slides: 20
Provided by: lesr155
Learn more at: http://www.gridpp.ac.uk
Category:
Tags: cernit | etienne | fabric | les | make | robertson | up

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: les robertson cernit 1


1
LHC Computing Grid Project Status
  • February 2002
  • www.cern.ch/lcg

2
LCG Status
  • Committee membership
  • Resource management
  • First RTAGs
  • Recruitment status
  • Trans-Atlantic inter-operability
  • Launching Workshop
  • Possible delay in the machine

3
Committee Membership
  • Project Overview Board
  • Chair CERN Director for Scientific Computing
  • experiment spokespersons
  • Representatives of countries/regions with Tier-1
    centres
  • France, Germany, Italy, Japan, United Kingdom,
    USA
  • 4 Representatives of countries/regions with
    Tier-2 centres from CERN Member States
  • process for selecting these not yet clear
  • ex officio IT Division Leader, SC2 chair,
    project leader
  • First meeting mid March

4
  • Software and Computing Committee (SC2)
  • Sets requirements, approves the strategy,
    monitors progress
  • Chair Matthias Kasemann (FNAL)
  • 2 coordinators from each LHC experiment
  • Technical Managers from centers in each region
    represented in the POB
  • UK John Gordon
  • France Denis Linglin, Germany Marcel Kunze,
    Italy Federico Ruggieri, Japan tba, USA
    Lothar Bauerdick
  • Ian Foster
  • ex officio IT and EP division leaders, project
    manager

5
Project Execution Board
  • Manages the resources, development and operation
    of the project
  • Chair Project Leader Les Robertson
  • Area managers
  • applications Torre Wenaus
  • regional centre strategy Mirco Mazzucato
  • data challenges grid deployment Themis
    Bowcock (starts February)
  • grid technology Fabrizio Gagliardi
  • computing fabric management Wolfgang von Rüden
  • Chief technical officer Dave Foster
  • one delegate from each experiment ALICE Alberto
    Masoni, ATLAS Gilbert Poulard, CMS Lucas Taylor,
    LHCb Philippe Charpentier
  • external members
  • Tony Doyle - GridPP
  • Mirco Mazzucato - INFN Grid
  • François Etienne - IN2P3 Grid
  • Ruth Pordes - iVDGL/PPDG
  • Miron Livny US Grid Computer Science

6
Resources
  • Four components
  • CERN personnel and materials (including those
    contributed by member and observer states)
  • directly assigned, managed and reported by the
    project
  • Experiment resources
  • may be committed by experiments within formal
    project plan (resources ? goals ? milestones)
  • Regional centres
  • commitments to take part in testbeds, data
    challenges
  • national coordination or individual RCs?
  • complex formal relationship funding
    agency/physics institution/experiment/Grid
    project/CERN
  • Formal agreements by MoU
  • Grid projects
  • suppliers of technology
  • formal agreements on deliverables?
  • who provides software maintenance, operations
    support?
  • Computing RRB (Resource Review Board)
  • Funding agencies, Experiments, CERN
  • High level oversight for the first three sources

7
Requirements Specification
  • Requirements Technical Assessment Group RTAG
  • short-lived, small, focused study group to
    prepare recommendations for the SC2
  • basic template
  • chair appointed by SC2
  • one member from each experiment
  • a few technical experts
  • interim report after one month
  • final report after two months

8
Persistency RTAG
  • Chair David Malon
  • Mandate Write the product specification for the
    Persistency Framework for Physics Applications at
    LHC.
  • Guidance
  • RTAG may decide to address all types of data, or
    may decide to postpone some topics for other
    RTAGS, once the components have been identified.
  • The RTAG should develop a detailed description at
    least for the event data management.
  • Issues of schema evolution, dictionary
    construction/storage, object and data models
    should be addressed
  • Makeup - Rene Brun (ROOT), Dirk Duellmann(IT/DB),
    Vincenzo Innocente (CMS), Pere Mato (LHCb), Fons
    Rademakers (ALICE)
  • Interim report at Launching Workshop

9
RTAG on Process for Managing Software
  • Chair Fons Rademakers
  • Mandate Define a process for managing LCG
    software
  • establish structure for organising software,
    managing versions, distribution
  • identify external software packages to be
    supported
  • identify common tools for software development,
    including configuration and release management
  • Makeup - Reps from each experiment, IT/API
  • Interim report at Launching Workshop

10
RTAG to review maths libraries
  • Chair Fred James
  • Mandate Review current situation with maths
    libraries
  • review the current usage and status of maths
    libraries used by LHC experiments including
    NAG, GSL, CLHEP, ROOT
  • make recommendations for adoption,
    discontinuation
  • propose improvements
  • Makeup - Reps from each experiment, IT/API, Root,
    ..

11
RTAG on Common Grid Requirements
  • Chair Federico Carminati
  • Mandate Develop a common set of Grid
    requirements use cases for LHC experiments
  • Guidance consider the possibility of defining a
    HEP Grid services layer to mask the details of
    the different grid implementations from
    applications
  • Makeup Reps from each experiment, one person
    from each of Datagrid, iVDGL
  • Approved at 8 February meeting

12
Recruitment Status
  • First GridPP recruitment round
  • gt 100 applicants
  • interviews in December in Swindon and at CERN
  • 12 offers, 1 refusal, 1 postponed
  • 3 fellows, 5 staff members, 2 employed by UK
    universities
  • 1 person started February
  • 4 applications 2 data management 3 systems, 1
    management
  • Second round being organised now
  • advertising next week
  • interviews April
  • EPSRC fellows
  • 6 foreseen
  • 1 selected starts March
  • advertising now for a new round of recruitment

13
Other recruitment
  • Going slowly pledges ? reality
  • Italy
  • recruitment under way for 5 fellows, 8 INFN
    people (4 to work closely with experiments)
  • Germany
  • recruitment started for 8 people
  • Israel
  • recruiting 1 or 2 engineers
  • Poland
  • 1 person starts in April
  • Switzerland
  • 1 network expert
  • 2 applications people
  • France 1 person starts April

14
Trans-Atlantic Inter-operability
  • Background
  • Incompatibilities between the testbeds of the
    Datagrid domain and the US domain
  • HENP Intergrid Collaboration Board
  • ad hoc committee of the managers of the Grid
    projects and some people from experiments
  • Joint Technical Board representatives of US,
    Europe and Japan co-chairs Peter Clarke, Ruth
    Pordes
  • LCG Strategy - achieve and demonstrate
    interoperability by end 2002
  • Three components
  • Develop a unified set of requirements for Grid
    services RTAG
  • Active participation of the LHC experiments
    through their data challenge programmes in
  • the Datagrid testbed and the US iVDGL-led testbed
  • the GRIT testbed when available see next bullet
  • Support the development deployment of a
    converged testbed GRIT

15
GRIT
  • Grid Reference Intercontinental Testbed (GRIT)
  • proposed by the HIJTB for discussion in Toronto
    (17feb)
  • develop compatibility/interoperability of the
    infrastructure for authentication, grid
    information, service discovery, … and basic
    services for job submission, cross-domain data
    movement, authorisation, ……
  • GRIT does not necessarily expect to select
    identical solutions on both sides of the
    Atlantic, but rather aims for interoperability of
    protocols, data formats, and services.
  • European funding for development within
    DataTAG US funding being identified

16
LCG GRIT Concerns
  • Long term goal is one set of middleware
  • Essential to have active participation by Grid
    applications experts from the LHC experiments in
    the definition and validation of the services
  • Importance of planning for deployment including
    finding resources
  • All Tier 1 sites and some Tier 2 sites must be
    included in deployment plan
  • Convergence with Datagrid technical programme

17
Launching Workshop
  • Goal Arrive at high level work-plan
  • Define scope and priorities
  • Understand detailed priorities for next 15-18
    months
  • Agree on coordination mechanisms
  • Establish collaborative spirit
  • Monday 11 March pm Plenary -
  • Status of computing in experiments, LCG, Overview
    of Grid projects, Regional Centres
  • Tuesday am Applications
    pm Fabrics
  • Wednesday am Grid Technology
    pm Grid Deployment
  • Thursday am Plenary General talks by
    non-HEP Grid specialists pm
    joint SC2/PEB meeting to prepare conclusions
  • Friday am Plenary Summaries, next steps
  • Friday pm - POB

18
Effects of Possible Delay in LHC Completion
LCG has two phases
  • Phase 1 2002-04
  • Development and prototyping
  • Funded by special contributions from member and
    observer states
  • Phase 2 2005-07
  • Installation and operation of the full world-wide
    initial production Grid
  • Costs (materials staff) included in the LHC
    cost to completion estimates
  • What happens if first physics is in 2007?

19
First physics in 2007
  • With fixed requirements, one year delay in
    building the production system implies
  • about CHF 30M reduction in costs at CERN (25 of
    the unfunded part)
  • 30 reduction in number of processors needed
  • one year gap between Phase I and Phase II ? OR
    extension of Phase I by one year ?
  • Looking at the cost implications of extending
    Phase I
  • manpower probably no significant change due to
    slow ramp-up
  • materials some savings (15)
  • but what does this mean for funding agencies?
About PowerShow.com