1. INFSO-RI-508833 Enabling Grids for E-sciencE www.eu-egee.org Status and evolution of the EGEE Project and its Grid Middleware By Frédéric Hemmer Middleware Manager CERN Geneva, Switzerland International Conference on Next Generation Networks Brussels, Belgium June 2, 2005

2. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 2 Contents The EGEE Project –Overview and Structure –Grid Operations –Middleware –Networking Activities –Applications  High Energy Physics  Biomedical Summary and Conclusions

3. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 3 EGEE goals Goal of EGEE: develop a service grid infrastructure which is available to scientists 24 hours-a-day The project concentrates on: –building a consistent, robust and secure Grid network that will attract additional computing resources –continuously improve and maintain the middleware in order to deliver a reliable service to users –attracting new users from industry as well as science and ensure they receive the high standard of training and support they need

4. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 4 EGEE EGEE is the largest Grid infrastructure project in Europe: 70 leading institutions in 27 countries, federated in regional Grids Leveraging national and regional grid activities ~32 M Euros EU funding for initially 2 years starting 1st April 2004 EU review, February 2005 successful Preparing 2 nd phase of the project – proposal to EU Grid call September 2005 Promoting scientific partnership outside EU

5. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 5 EGEE Geographical Extensions EGEE is a truly international under-taking Collaborations with other existing European projects, in particular:  GÉANT, DEISA, SEE-GRID Relations to other projects/proposals:  OSG: OpenScienceGrid (USA)  Asia: Korea, Taiwan, EU-ChinaGrid  BalticGrid: Lithuania, Latvia, Estonia  EELA: Latin America  EUMedGrid: Mediterranean Area  … Expansion of EGEE infrastructure in these regions is a key element for the future of the project and international science

6. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 6 EGEE Activities 48 % service activities (Grid Operations, Support and Management, Network Resource Provision) 24 % middleware re-engineering (Quality Assurance, Security, Network Services Development) 28 % networking (Management, Dissemination and Outreach, User Training and Education, Application Identification and Support, Policy and International Cooperation) Emphasis in EGEE is on operating a production grid and supporting the end-users

7. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 7 EGEE Activities Emphasis in EGEE is on operating a production grid and supporting the end-users

8. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 8 Computing Resources – April 2005 Country providing resources Country anticipating joining 131 sites, 30 countries  >12,000 cpu  ~5 PB storage

9. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 9 Infrastructure metrics Countries, sites, and CPU available in EGEE production service Countries, sites, and CPU available in EGEE production service Regioncoun- tries sitescpu M6 (TA) cpu M15 (TA) cpu actual CERN0190018001841 UK/Ireland21910022002398 France184008951172 Italy1215536792164 South East516146322159 South West213250 498 Central Europe510385730629 Northern Europe242002000427 Germany/Switzerland2101004001733 Russia1950152276 EGEE-total211113084942811297 USA13--555 Canada16--316 Asia-Pacific68--394 Hewlett-Packard13--172 Total other920--1437 Grand Total30131--12734 EGEE partner regions Other collaborating sites

10. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 10 Service Usage VOs and users on the production service –Active HEP experiments:  4 LHC, D0, CDF, Zeus, Babar –Active other VO:  Biomed, ESR (Earth Sciences), Compchem, Magic (Astronomy), EGEODE (Geo-Physics) –6 disciplines –Registered users in these VO: 600 –In addition to these there are many VO that are local to a region, supported by their ROCs, but not yet visible across EGEE Scale of work performed: –LHC Data challenges 2004:  >1 M SI2K years of cpu time (~1000 cpu years)  400 TB of data generated, moved and stored  1 VO achieved ~4000 simultaneous jobs (~4 times CERN grid capacity) Number of jobs processed/month

11. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 11 Grid Operations The grid is flat, but Hierarchy of responsibility –Essential to scale the operation CICs act as a single Operations Centre –Operational oversight (grid operator) responsibility –rotates weekly between CICs –Report problems to ROC/RC –ROC is responsible for ensuring problem is resolved –ROC oversees regional RCs ROCs responsible for organising the operations in a region –Coordinate deployment of middleware, etc CERN coordinates sites not associated with a ROC CIC RC ROC RC ROC RC ROC RC ROC OMC RC - Resource Centre ROC - Regional Operations Centre CIC – Core Infrastructure Centre

12. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 12 Grid monitoring –GIIS Monitor + Monitor Graphs –Sites Functional Tests –GOC Data Base –Scheduled Downtimes –Live Job Monitor –GridIce – VO + fabric view –Certificate Lifetime Monitor Operation of Production Service: real-time display of grid operations Accounting information Selection of Monitoring tools:

13. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 13 LCG Deployment Schedule

14. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 14 LCG Service Challenges “Service Challenge 2” –Throughput test from LCG Tier-0 to LCG Tier-1 sites –Started 14 th March Set up Infrastructure to 7 Sites –NL, IN2P3, FNAL, BNL, FZK, INFN, RAL 100MB/s to each site –500MB/s combined to all sites at same time –500MB/s to a few sites individually Goal : by end March’05, sustained 500 MB/s at CERN

15. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 15 SC2 met its throughput targets >600MB/s daily average for 10 days was achieved - Midday 23 rd March to Midday 2 nd April –Not without outages, but system showed it could recover rate again from outages –Load reasonable evenly divided over sites (give network bandwidth constraints of Tier-1 sites)

16. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 16 Service Challenge 3 Throughput phase –2 weeks sustained in July 2005 –Primary goals:  150MB/s disk – disk to Tier1s;  60MB/s disk (T0) – tape (T1s) –Secondary goals:  Include a few named T2 sites (T2 -> T1 transfers)  Encourage remaining T1s to start disk – disk transfers Service phase –September – end 2005  Start with ALICE & CMS, add ATLAS and LHCb October/November  All offline use cases except for analysis  More components: WMS, VOMS, catalogs, experiment-specific solutions –Implies production setup (CE, SE, …)

17. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 17 EGEE Activities Emphasis in EGEE is on operating a production grid and supporting the end-users

18. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 18 Future EGEE Middleware - gLite Application requirements http://egee-na4.ct.infn.it/requirements/http://egee-na4.ct.infn.it/requirements/ Intended to replace present middleware (LCG-2) Developed mainly from existing components Aims to address present shortcomings and advanced needs from applications Regular, iterative updates for fast user feedback Makes use of web-services where currently feasible Globus 2 basedWeb services based gLite-2gLite-1LCG-2LCG-1

19. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 19 Architecture & Design Design team including representatives from Middleware providers (AliEn, Condor, EDG, Globus,…) and Operations, including US partners produced middleware architecture and design. Takes into account input and experiences from applications, operations, and related projects –Focus on medium term (few months) and commonalities with other projects (e.g. OSG) –Effective exchange of ideas, requirements, solutions and technologies –Coordinated development of new capabilities –Open communication channels –Joint deployment and testing of middleware –Early detection of differences and disagreements The 2 nd release of gLite (v1.1) made in May’05 –http://cern.ch/glite/packages/R1.1/R20050430/default.asphttp://cern.ch/glite/packages/R1.1/R20050430/default.asp –http://cern.ch/glite/documentationhttp://cern.ch/glite/documentation gLite is not “just” a software stack, it is a “new” framework for international collaborative middleware development. Much has been accomplished in the first year. However, this is “just” the first step.

20. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 20 gLite Services in Release 1 Software stack and origin (simplified) Computing Element –Gatekeeper (Globus) –Condor-C (Condor) –CE Monitor (EGEE) –Local batch system (PBS, LSF, Condor) Workload Management –WMS (EDG) –Logging and bookkeeping (EDG) –Condor-C (Condor) Information and Monitoring –R-GMA (EDG) Storage Element –glite-I/O (AliEn) –Reliable File Transfer (EGEE) –GridFTP (Globus) –SRM: Castor (CERN), dCache (FNAL, DESY), other SRMs Catalog –File/Replica & Metadata Catalogs (EGEE) Security –GSI (Globus) –VOMS (DataTAG/EDG) –Authentication for C and Java based (web) services (EDG) Now doing rigorous scalability and performance tests on pre-production service

21. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 21 Software Process JRA1 Software Process is based on an iterative method It comprises two main 12-month development cycles divided in shorter development-integration-test-release cycles lasting 1 to 4 weeks The two main cycles start with full Architecture and Design phases, but the architecture and design are periodically reviewed and verified. The process is documented in a number of standard documents: –Software Configuration Management (SCM) Plan –Test Plan –Quality Assurance Plan –Developer’s Guide

22. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 22 Release Process DevelopmentIntegrationTesting Software Code Deployment Packages Integration Tests FailPass Fix Functional Tests Testbed Deployment Fail Pass Installation Guide, Release Notes, etc

23. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 23 Bug Counts and Trends May 18, 2005 # Defects/KLOC: 2.01

24. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 24 gLite: What’s next? Focus and Priority is –Bug Fixing –Support to Service Challenge 3  File Transfer Service (FTS) New planned features in 1.2 –VOMS 1.5 (Oracle support) –CE: Condor support (in addition to PBS/LSF) –WMproxy (Web Services Interface including bulk job submission) –Service discovery interface to BDII –File Transfer Service improvements –R-GMA aligned with the LCG-2 version Beyond 1.2 –DGAS accounting system –Job Provenance –Globus Workspace Services –Harmonization of Security Models –Integration with Service Discovery

25. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 25 EGEE Activities Emphasis in EGEE is on operating a production grid and supporting the end-users

26. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 26 Outreach & Training Public and technical websites constantly evolving to expand information available and keep it up to date 3 conferences organised –~ 300 @ Cork, ~ 400 @ Den Haag, ~500 @ Athens Pisa 4th project conference 24-28 October ’05 More than 70 training events (including the GGF grid school) across many countries –~1000 people trained  induction; application developer; advanced; retreats –Material archive with more than 100 presentations Strong links with GILDA testbed and GENIUS portal developed in EU DataGrid

27. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 27 Deployment of applications Pilot applications –High Energy Physics –Biomed applications http://egee-na4.ct.infn.it/biomed/applications.html Generic applications – Deployment under way –Computational Chemistry –Earth science research –EGEODE: first industrial application –Astrophysics With interest from –Hydrology –Seismology –Grid search engines –Stock market simulators –Digital video etc. –Industry (provider, user, supplier) PilotNew

28. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 28 HEP (ATLAS) Utilisation ~660K jobs total in (LCG, Nordugrid, US Grid3) ~400 kSI2k years of CPU In latest period average ~7K jobs/day with ~5K in LCG 12000 6000

29. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 29 Bioinformatics GPS@: Grid Protein Sequence Analysis –NPSA is a web portal offering proteins databases and sequence analysis algorithms to the bioinformaticians (3000 hits per day) –GPS@ is a gridified version with increased computing power –Need for large databases and big number of short jobs xmipp_MLrefine –3D structure analysis of macromolecules from (very noisy) electron microscopy images –Maximum likelihood approach for finding the optimal model –Very compute intensive Drug discovery –Health related area with high performance computation need –An application currently being ported in Germany (Fraunhofer institute)

30. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 30 Medical imaging GATE –Radiotherapy planning –Improvement of precision by Monte Carlo simulation –Processing of DICOM medical images –Objective: very short computation time compatible with clinical practice –Status: development and performance testing CDSS –Clinical Decision Support System –knowledge databases assembling –image classification engines widespreading –Objective: access to knowledge databases from hospitals –Status: from development to deployment, some medical end users

31. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 31 Medical imaging SiMRI3D –3D Magnetic Resonance Image Simulator –MRI physics simulation, parallel implementation –Very compute intensive –Objective: offering an image simulator service to the research community –Satus: parallelized and now running on EGEE resources gPTM3D –Interactive tool for medical images segmentation and analysis –A non gridified version is distributed in several hospitals –Need for very fast scheduling of interactive tasks –Objectives: shorten computation time using the grid –Status: development of the gridified version being finalized

32. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 32 Status of Biomedical VO PADOVA BARI 20 resource centres  CEs: >2000 CPUs  SEs: >20TB disk 4 RBs: CNAF, IFAE, LAPP, UPV RLS, VO LDAP Server: CC-IN2P3 15 RBs 2 RLS 1 LDAP Server

33. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 33 Grid conclusions e-Infrastructures deployment creates a powerful new tool for science – as well as applications from other fields Investments in grid projects and e-Infrastructure are growing world-wide Applications are already benefiting from Grid technologies Open Source is the right approach for publicly funded projects and necessary for fast and wide adoption Europe is strong in the development of e-Infrastructure also thanks to the initial success of EGEE Collaboration across national and international programmes is very important

34. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 34 Summary EGEE is the first attempt to build a worldwide Grid infrastructure for data intensive applications from many scientific domains A large-scale production grid service is already deployed and being used for HEP and BioMed applications with new applications being ported Resources & user groups are expanding A process is in place for migrating new applications to the EGEE infrastructure A training programme has started with many events already held “next generation” middleware is being tested (gLite) First project review by the EU successfully passed in Feb’05 Plans for a follow-on project are being prepared

35. Enabling Grids for E-sciencE INFSO-RI-508833 Next Generation Networks - June 2, 2005 35 Contacts EGEE Web Site http://www.eu-egee.org How to join http://public.eu-egee.org/join/ gLite Web Site http://www.glite.org EGEE Project Office project-eu-egee-po@cern.ch