ed66069f81aa3cc55cb881bb3fcc7aae.ppt
- Количество слайдов: 34
LCG-France Vincent Breton, Eric Lançon and Fairouz Malek, CNRS-IN 2 P 3 and LCG-France ISGC Symposium Taipei, March 27 th 2007
LCG-France Project • Goals § Setup, develop and maintain a LCG Tier-1 and an Analysis Facility at CC-IN 2 P 3 § Promote the creation and coordinate the integration of Tier-2/Tier-3 french sites into the LCG collaboration • Funding § national funding for tier-1 and AF § Tier-2 s and tier-3 s funded by universities, local/regional governments, hosting laboratories, … • Organization § Started in June 2004 § Scientific and technical leaders appointed, management board (executive) and overview boards in place since then V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 2
LCG-France sites V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 3
LCG France contribution (October 2005 - September 2006) • EGEE « CPU accounting » per EGEE region § French contribution includes LCG-France sites V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 4
sites contribution to LCG France (September 2005 - October 2006) • « CPU accounting » per site for all EGEE Virtual Organizations V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 5
Tier-1 Contribution • Planned contribution of LCG-France Tier-1 § % of required resources for all tier-1 s in 2008 Source: Comparison of New Requirements with Current Pledges – 24/10/2006 V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 6
Tier-1 Contribution (cont. ) • Planned contribution of LCG-France tier-1 § % of required resources in all tier-1 s in 2008 (to be finalized) V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 7
Tier-2 s Contribution V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 8
Tier-1 Planned Evolution Increase rate over the period 2006 -2010: CPU: x 17 DISK: x 16 MSS: x 18 V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 9
Tier-2 s Planned Evolution Roughly equivalent to the planned Tier-1 CPU capacity the same year V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 43% of the planned Tier-1 disk capacity the same year 10
Tier-3 s: Planned Evolution Data for IPNL (Lyon) are not included V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 11
Tier-2/Tier-3 Activities • Coordination of LCG-France tier-2/tier-3 technical activities officially set up in April 2006 § Frédérique Chollet is leading the group § Collaboration tools in place w Mailing list, wiki pages, regular video-conference meetings • Activities § Very active in the Quattor working group w Used by most of the LCG-France sites § Network-level and SRM-level data transfer tests from and to tier-1 w Including associated foreign sites (more on this later) § Meetings held with several potential hardware providers w Sharing of technical and commercial information (hardware evaluation results, commercial conditions, etc. ) V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 12
Tier-2/Tier-3 Activities (cont. ) • In close contact with some foreign associated tier-2 s § Europe w Belgium CMS Tier-2 w Romanian Federation ATLAS Tier-2 § Asia w IHEP China - ATLAS and CMS Tier 2 w ICEPP Japan - ATLAS Tier 2 V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 13
Tier-1: site overview Mon. Box 4 Sites VOMS 4 VOs Site BDII LFC Local 4 VOs LHC LFC Central Biomed FTS 4 VOs LHC Gridftp BQS Service global Storage Element XFS Computing Element Gridftp HPSS DCACHE Anastasie Service régional/fédéral Service local WN WN Calcul V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, Stockage 14 SRM Courtesy of Pierre Girard Système d’informatio n de la grille VO Box VO LHC V OBox VO LHC
Tier-1: site overview (cont. ) • Operating also several grid services for non. LHC VOs V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 15
Tier-1 Contribution in 2006 • CPU time contributed by the french tier-1 in 2006 Source: WLCG Accounting Report Tier-1 Centres + CERN § % of CPU time (grid and non-grid) used by the experiments in all the tier-1 s V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, The CC-IN 2 P 3 contribution to the global effort in 2006 was 10% of the total CPU used by the 4 experiments in all the tier-1 s. 16
Tier-1 Contribution in 2006 (cont. ) • CPU utilisation by LHC experiments at all the tier-1 s and at CC-IN 2 P 3 All Tier-1 s (does not include non-grid usage of some sites) Source: http: //www 3. egee. cesga. es/gridsite/accounting/CESGA/tier 1_view. html V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 17 CC-IN 2 P 3 (grid and non-grid)
Tier-1: grid vs. non-grid usage • Site usage (grid vs. non -grid) greatly varies from one experiment to another § Both in terms of consumed capacity and number of jobs V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 18
Tier-1: efficiency (CPU time vs. wallclock) Measurement error. V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 19
Tier-1: CPU planned vs. actual consumption V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 20
Tier-1: CPU capacity delivered Several service interruptions in august and september due to problems with the cooling or power infrastructure 4 days-long scheduled complete shutdown of the site for replacing some central electric equipement V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 21
Tier-1: storage delivered • Disk storage capacity § Delivered 34% (180 TB out of 520 TB planned) § More on this later • Tape storage capacity § Installed capacity (as planned) of 535 TB (of which 73% was actually used) V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 22
Tier-1: data transfer exercises • CERN → CC-IN 2 P 3 (disk) § April 2006 Targ et: 200 MB/s ec • CERN → CC-IN 2 P 3 (MSS) § April 2006 V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, Target: 75 MB/se c 23
Source: http: //lcg. web. cern. ch/LCG/MB/availability/site_reliability. pdf Tier-1: site availability V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 24
Tier-1: capacity increase in 2006 • CPU § +265 worker nodes (IBM, dualprocessor dual-core AMD Opteron 275, 2. 2 GHz, 2 GB/core, 290 GB internal disk) § Theoretical power: 1573 SI 2000 per core w Total: 1, 6 M SI 2000 w Observed power with typical applications is ~30% less than theoretical • Disk storage § +400 TB of rack-mounted Sun Fire X 4500 (aka Thumper) V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 25
Tier-1: capacity increase in 2006 (cont. ) • Tape storage § Call for tender for a new cartridge library § Selected Sun/Storage. Tek SL 8500 w 30 T 10000 drives w 10 LTO-3 drives § Will progressively replace the current one w 6 silos § Installation started in January w Expected to be finished by end of april 2007 V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 26
Tier-1: capacity increase in 2006 (cont. ) • Databases § Reconfiguration of Oracle cluster w Extensible hardware architecture § +1 TB added to the dedicated SAN (2 TB total) § +3 front-end database servers (5 total) w 2 of them will share the load of the LHC experiments • International connectivity § Dedicated link CC-IN 2 P 3↔CERN 10 Gbps § 2 x 1 Gbps links CC-IN 2 P 3 ↔ Fermilab V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 27
Hardware procurement • Procurement process (evaluation, publication, selection) is more or less under control § Delivery delays are not! § In 2006, we suffered delivery delays of several months for some equipment w A fraction of the equipment is still not delivered! • Procurement of equipment is an issue § Several constraints: space in the machine room, budget constraints, delivery delays, requested availability, … V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 28
Facility Upgrade • Major effort for upgrading the electric and cooling infrastructure of the site § Currently reaching the limits of the installation § When the current works will be finished (april 2007) Average total electrical power monthly consumption V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 29 Courtesy of Dominique Boutigny w from 500 k. W to 1000 k. W usable for computing equipement
Facility Upgrade (cont. ) • Scheduled 4 days-long complete shutdown of the site in December 2006 for replacing central electric equipment § Vital services (network equipment, mail servers, web servers, Oracle, FTS, LFCs, VOMS, …) were kept alive by ad hoc means) w Extensive use of virtual machines § Others services have been switched to partner sites w CIC Portal was hosted by CNAF during the shutdown and switched back to CC-IN 2 P 3 afterwards w Failover procedure tested in real conditions V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 30
Plans for 2007 • Tier-1 § Consolidate current grid services and integrate them into « normal » operations w Works towards the stability desired not only by the experiments but by the people operating the services § Increase network bandwidth with tier-2 s and backup link to other tier 1 s through FZK § Increase the rythm of the new machine room building project planned for 2009 • Tier-2 s/Tier-3 s § Improve availability of the sites § Keep exercising the data transfer infrastructure • All § Make sure site administrators understand the ways data will be accessed (!) V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 31
Conclusions • Participating sites are very motivated to contribute to this project… § … but it is harder than most of us expected • Ramp up plans of sites are rather aggressive § Several constraints don’t really make our life easier • Operating the grid services in their current status is complex and requires (highly competent and motivated) people V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 32
For more information • LCG-France website http: //lcg. in 2 p 3. fr • LCG-France T 2 -T 3 Technical coordination wiki page: http: //lcg. in 2 p 3. fr/wiki/index. php/T 2 T 3 • CC-IN 2 P 3: http: //cc. in 2 p 3. fr • LCG-France Tier-1 resource planning § https: //edms. in 2 p 3. fr/document/I-004736 • LCG-France Tier-2 s resource planning § https: //edms. in 2 p 3. fr/document/I-008142 V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 33
Aknowledgments • Thanks to the people that contributed material to this talk § Most of the slides are taken from Fabio Hernandez talk at WLCG Collaboration Workshop (CERN, January 22 nd 2007) § Special thanks to Eric Lançon and Fairouz Malek V. Breton, E. Lançon and F. Malek, ISGC symposium, March 27 th 2007, 34
ed66069f81aa3cc55cb881bb3fcc7aae.ppt