1. LCG and HEPiX Ian Bird LCG Project - CERN HEPiX - FNAL 25-Oct-2002

2. CERN Ian.Bird@cern.ch 2 What is LCG? Why is it relevant to HEPiX?

3. CERN Ian.Bird@cern.ch 3 LCG Project Goals  applications - tools, frameworks, environment, persistency  computing system  global grid service  cluster  automated fabric  collaborating computer centres  grid  CERN-centric analysis  global analysis environment  central role of data challenges Goal – Prepare and deploy the LHC computing environment This is not another grid technology project – it is a grid deployment project

4. CERN Ian.Bird@cern.ch 4 LCG Level 1 Milestones proposed to LHCC M1.1 - June 03First Global Grid Service (LCG-1) available M1.2 - June 03Hybrid Event Store (Persistency Framework) available for general users M1.3a - November 03LCG-1 reliability and performance targets achieved M1.3b - November 03Distributed batch production using grid services M1.4 - May 04Distributed end-user interactive analysis from “Tier 3” centre M1.5 - December 04“50% prototype” (LCG-3) available M1.6 - March 05Full Persistency Framework M1.7 - June 05LHC Global Grid TDR

5. CERN Ian.Bird@cern.ch 5 GTA Common Applications Deployment Fabric Experiments Regional Centres GDB GLUE HEPCAL CERN PPDG iVDGL (VDT) EDG NorduGrid AliEn GriPhyN Globus Grid Projects LCG and its interactions

6. CERN Ian.Bird@cern.ch 6 Multi-dimensional problem  Regional Centres:  Host one or more experiments  Different RC’s deploy different grid middleware in existing testbeds  Have different operational and security policies  Experiments:  Use middleware from various grid projects  Run at many regional centres  Provide applications that rely on specific middleware  Grid projects:  Provide middleware – that does not often (yet) interoperate  Starting to collaborate on common solutions and interoperability  The Deployment area of LCG ties these all together

7. CERN Ian.Bird@cern.ch 7 Grid Deployment – goals of LCG-1  Production service for Data Challenges in 2H03 & 2004  Focused on batch production work  Experience in close collaboration between the Regional Centres  Should have wide enough participation to understand the issues, but not too many initially  Learn how to maintain and operate a global grid  Focus on a production-quality service and all that implies  Robustness, fault-tolerance, predictability, and supportability take precedence over functionality  But – minimum functionality to be of value  This requires:  a middleware support group with integration, certification, testing, packaging etc. responsibilities  A support structure  LCG should be integrated into the sites’ physics computing services – should not be something apart  This requires coordination between participating sites in:  Policies and collaborative agreements  Resource planning and scheduling  Operations  Support

8. CERN Ian.Bird@cern.ch 8 What might LCG-1 look like?  User’s perspective: - requires  Functionality adequate to provide advantage over not using distributed model  Straightforward to use –  Well defined services  Advice on how to use the system  Help with problems  Failures should be understandable  Ability to determine status of jobs and data  Sites’ perspective:  Integrated into computer centre/IT (inc. security) infrastructures  Able to support service  Able to allocate and manage resources – local autonomy where needed  Overall service perspective:  Performance and problem monitoring  Accounting  Etc.

9. CERN Ian.Bird@cern.ch 9  LCG has to build the “virtual computer centre” (= LHC computing environment)  With all that is expected from a production service  User support  Operations group  “Account” management  Security  Fabric management  Etc..  Except this is now distributed across many countries and continents  Requires agreements, collaboration, and coordination  At all levels: management, system managers, user support, etc.

10. CERN Ian.Bird@cern.ch 10 Grid Operation User Network Operations Centre Local operation Local user support Grid Operations Centre Call Centre Local site Grid operations Grid information service Virtual Organisation Grid logging & bookkeeping queries monitoring & alarms corrective actions

11. CERN Ian.Bird@cern.ch 11 Deployment Summary  Deploy middleware to support essential functionality, but goal is to evolve and incrementally add functionality  Added value is to robustify, support and make into a 24x7 production service  How?  Certification & test procedure – tight feedback to developers  must develop support agreements with grid projects to ensure this  Define missing functionality – require from providers  Provide documentation and training  Provide missing operational services  Provide a 24x7 Operations and Call Centre  Guarantee to respond  Single point of contact for a user  Make software easy to install – facilitate new centres joining

12. CERN Ian.Bird@cern.ch 12 LCG Strategy  Develop as little as possible  Use existing middleware, tools and software  Pressure developers to provide missing functionality  Negotiate support agreements  Leverage existing experience  Various data grid projects and testbeds  Teragrid, interoperability demonstrations, GGF – production grids area  Actively encourage collaboration and coordination

13. CERN Ian.Bird@cern.ch 13 Grid Deployment Teams – the plan common applications s/w Trillium - US grid middleware DataGrid middleware certification, build & distribution LCG infrastructure coordination & operation user support call centre fabric operation regional centre A grid operation fabric operation regional centre B fabric operation regional centre Y fabric operation regional centre X … suppliers’ integration teams provide tested releases LCG HEPiX interests

14. CERN Ian.Bird@cern.ch 14 Coordination & Collaboration  There are many opportunities for common solutions, which should be actively pursued  HICB – JTB, existing & proposed new collaborative activities  GLUE  Schema definitions & interoperability work  Validation and Test Suites  Distribution and Meta-Packaging  Interoperable distribution and configuration utilities identified as a definite need by all the recent trans-Atlantic demonstration and validation work.  Support for this group comes from: –LCG, EDG, EDT, Trillium, DataTAG  Security czars  Already talking to address grid issues  GGF  Production grids  AAA  Etc.  LCG – grid deployment board, etc.

15. CERN Ian.Bird@cern.ch 15 Summary of Issues that might be addressed by HEPiX/LCCWS  I know many of these are discussed by a plethora of grid projects and offshoots, but remember, more than ever before we all have to work together coherently to make a grid work:  Grid operations centre: Teragrid, iVDGL  User support –  distributed helpdesk/call centre: iVDGL, Teragrid, Nordic grid collabs, GGF production grids area  Helpdesk tools  Certification process for operating environments  Upgrade procedures  Configuration management  Joint OS version certification  Packaging, installation – inc applications  User management  Security etc.  Fabric management (see LCCWS)  Etc.

16. CERN Ian.Bird@cern.ch 16 Proposal  HEPiX is already (a lot of) the right people  Already, or soon to be, deploying LCG and other grids in their computer centres  Keep LCCWS associated with HEPiX  Add a Grid Coordination/LCG interest group – like HEPNT or Storage  To address themes and issues of common interest  Encourage new people to attend  Line up specific talks by selected people to address issues and to propose activities to follow on  We need to solve the problems – not just talk about them  Needs a coordinator & agenda to make sure this happens –  Volunteers?