les robertson cernit 1

1
CERN-C-RRB-2004-011

• The LHC Computing Grid Project
• Status Report
• Computing RRB 27 April 2004
• Les Robertson LCG Project Leader
• CERN European Organization for Nuclear Research
• Geneva, Switzerland
• les.robertson_at_cern.ch

2
Outline

• Supplementary information to the presentation in
  Mondays session
• The LCG grid service
• EGEE and Grid3/OSG
• Milestone chart
• Requirements for Phase 2
• Staffing
• Conclusion

3
Hewlett Packard to provide Tier 2-like services
for LCG, initially in Puerto Rico
4
GOC
Secure Database Management via HTTPS / X.509
GOC GridSite MySQL
Monitoring
Resource Centre Resources Site Information EDG,
LCG-1, LCG-2,
bdii
ce
se
rb
RC
5
(No Transcript)
6
Global Grid User Support - GridKa

• Grid-wide service for Grid specific problems
• Problem tracking and management
• Grid help desk
• Interacts with
• Experiment User Support
• Expert users
• Site user support centres (help desks)
• Grid Operations Centres (RAL, Taipei)
• Grid infrastructure coordination centre at CERN
• Started with support for GridKa Users (02 Oct.
  2003)
• Wider support starting now

7
LCG-2 Support Agreements
Most of these services are covered only
by informal agreementsalthough
substantialresources may beinvolved
8
LCG-2 and Next Generation Middleware
LCG-2
Next Generation
prototype
product development
mainline service

• LCG-2 will be the main service for the 2004 data
  challenges
• This will provide essential experience on
  operating and managing a global grid service
  and will be supported and developed to achieve
  these goals
• Target is to establish a base (fallback) solution
  for early LHC years
• The LCG-2 middleware will be maintained and
  developed until the new generation has proven
  itself
• This puts pressure on the new middleware
  developers to replace and add components rather
  than require a complete change

9
Service Challenges preparing for LHC Data Taking

• In parallel with experiments data challenges, we
  will start a series of Service Challenges
• Learn how to operate the grid reliably for
  extended periods
• Develop and operate data distribution services
• Mass-store to mass store
• Network performance, architecture
• Initially prototyping then pilot services
• First discussions with regional centre staff at
  HEPiX workshop at end of May
• Also discussing collaboration with GEANT/NRENs

10
Toward the US Open Science Grid

• Building partnerships on US Grid infrastructure
  for LHC and other sciences
• LHC application driving this effort, Grid3 is a
  great initial step
• Federate US resources with the LCG, the EGEE and
  other national and international
  Grids
• US LHC experiment projects, regional centers,
  universities and Grid projects
• formulated a roadmap towards the Open Science
  Grid

11
LCG EGEE Grid3/OSG

• LCG must prepare for a long-term operation
• it must include all the computing resources
  available for LHC Europe, US, Asia,..
• it must make it as easy as possible for
  experiments to use these resources
• This will require some form of inter-operability
  between grids that use different technologies,
  standards
• So, for the foreseeable future, - LCG will
  have a coordinating role as well as an operating
  role
• EGEE and LCG (operating role) are closely
  integrated by design
• Operations same management, shared
  infrastructure
• Middleware EGEE starts with the LCG-2 package
  - and we evolve together
• HEP Applications activity in EGEE - embedded in
  LCG distributed analysis activity
• Representation in each others management
  committees
• We still need to work out how the service is
  governed - policy, strategy, etc. LCG does this
  with a Grid Deployment Board national
  representatives VOs
• We are actively discussing how we synchronise
  with Grid3/OSG in the US

12
(No Transcript)
13
Phase 2 Planning
14
RAL
small centres
Santiago
desktops portables
Tier-2
Weizmann
Tier-1
IN2P3
FNAL

• LHC Computing Model (simplified!!)
• Tier-0 the accelerator centre
• Filter ? raw data
• Reconstruction ? summary data (ESD)
• Record raw data and ESD
• Distribute raw and ESD to Tier-1
• Tier-1
• Permanent storage and management of raw, ESD,
  calibration data, meta-data, analysis data and
  databases? grid-enabled data service
• Data-heavy analysis
• Re-processing raw ? ESD
• National, regional support

CNAF
FZK
PIC
ICEPP
BNL
online to data acquisition process -- high
availability -- managed mass storage --
long-term commitment
15
RAL
small centres
Santiago
desktops portables
Tier-2
Weizmann
Tier-1
IN2P3
FNAL
CNAF
FZK

• Tier-2
• Well-managed disk storage grid-enabled
• Simulation
• End-user analysis batch and interactive
• High performance parallel analysis (PROOF)

PIC
ICEPP
BNL
16
Estimates prepared as input to the MoU Task
Force Computing models under active development
17


Does not include traffic between Tier-1s
and with Tier-2s
18
Planning for the Phase 2 Grid

• The Phase 2 services at least for Tier-0 and
  Tier-1 centres must be in operation by September
  2006
• Acquisition process for the scale of computing
  required is very long in some centres -- and is
  starting now at CERN
• ? Tier-0, Tier-1 centres will be well into their
  planning and acquisition process before the MoU
  is signed and the TDR completed (July 2005)

Prototype services ? pilot services
19
LCG Staffing for the Common Activities at CERN
Experiments
EGEE Phase 1
External funding LCG Phase 1
Funding not yetidentified
20
Conclusion

• The first products of the common applications
  project are in use by the experiments
• The nucleus of a grid infrastructure for LHC is
  in place
• Beginning now to learn about data handling,
  operation, management
• Phase 1 of the project has been possible because
  of your contributions to the common activities at
  CERN as well as your investments in developments
  and services in regional centres.
• LCG must exploit short-term projects (like EGEE)
  but remembering that LHC has long-term
  requirements
• Agreement on how to incorporate US grid resources
  is an important goal for the coming months
• Estimates are available for the capacity and
  networking requirements for Tier-1 and Tier-2
  centres the numbers and the computing model
  will evolve, but detailed planning for these
  centres can begin
• Important that regional centres have resources to
  participate in Service Challenges to learn and
  validate their operation over the next two years
• Funding for the staffing of the common activities
  at CERN remains a major concern for the project