CMS Tier 1 at JINRV.V. Korenkovfor JINR CMS Tier-1 TeamJINR XXIV International Symposium on Nuclear Electronics & Computing, NEC2013 2013, September 13
Outline • CMS Grid structure • role of Tier-1s • CMS Tier-1s • CMS Tier-1 in Dubna • History and Motivations (Why Dubna?) • Network infrastructure • Infrastructure and Resources • Services and Readiness • Staffing • Milestones • Conclusions
Tier Structure of GRID Distributed Computing:Tier-0/Tier-1/Tier-2 Tier-0 (CERN): • accepts data from the CMS Online Data Acquisition and Trigger System • archives RAW data • the first pass of reconstruction and performs Prompt Calibration • data distribution to Tier-1 Tier-1 (11 centers): • receives a data from the Tier-0 • data processing (re-reconstruction, skimming , calibration etc) • distributes data and MC to the other Tier-1 and Tier-2 • secure storage and redistribution for data and MC Tier-2 (>200 centers): • simulation • user physics analysis 3 Ian.Bird@cern.ch
Tier1 center The Federal Target Programme Project: «Creation of the automated system of data processing for experiments at the LHC of Tier-1 level and maintenance of Grid services for a distributed analysis of these data» Duration:2011 – 2013 • March 2011 - Proposal to create the LCG Tier1 center in Russia(official letter by Minister of Science and Education of Russia A. Fursenko has been sent to CERN DG R. Heuer): • NRC KI for ALICE, ATLAS, and LHC-BLIT JINR (Dubna) for the CMS experiment September 2012 – Proposal was reviewed by WLCG OB and JINR and NRC KI Tier1 sites were accepted as a new “Associate Tier1” Full resources - in 2014 to meet the start of next working LHC session.
Why in Russia? Why Dubna?
In frames of the RDIG project (a participant of the WLCG/EGEE projects), a grid-infrastructure excepted by LHC experiments has been successfully launched as a distributed cluster RuTier2 (Russian Tier2) and JINR cluster JINR-LCG2 is the main one in RDIG as to its performance. JINR-LCG2 ~40% of CPU time in RDIG for 2011-2013
JINR Central Information and Computing Complex (CICC) LocalJINR users (no grid) Jobs run by JINR Laboratories and experiments executed at CICC January - September 2013. Grid users (WLCG) JINR-LCG2 Normalised CPU time by LHC VOs. January - September 2013. More than 3 million jobs run Total normalised CPU time – 20 346 183 kSI2K-hours http://lit.jinr.ru/view.php?var1=comp&var2=ccic&lang=rus&menu=ccic/menu&file=ccic/statistic/stat-2013 http://accounting.egi.eu/
CMS Computing at JINR • the first RDMS CMS web-server (in 1996); • full-scale CMS software infrastructure support • since 1997 • JINR CMS Tier2 center is one of • the most reliable and productive • CMS Tier2 centers worldwide • (in the top ten best) the most • powerful RDMS CMS Tier2 center • CMS Regional Operation Center • are functioning in JINR since 2009 • The core services needed for WLCG Tier-1 are • computing service, a storage service, information service. • The primary Tier-1 tasks can be divided into • recording raw data from CERN and storing them on tape; • recording processed data from CERN and storing them on disk; • providing data to other Tier-1 / Tier-2; • reprocessing raw data; • event simulation calculations.
Russia Normalized CPU time per SITE and VO (2012-2013) All VO Russia - 409,249,900 JINR - 183,008,044 CMS Russia - 112,025,416 JINR - 67,938,700 (61%) 10
Infrastructure and Facilities
JINR CMS Tier-1 progress • Disk & server installation and tests: done • Tape system installation: done • Organization of network infrastructure and connectivity to CERN via GEANT:done • Registration in GOC DB and APEL: done • Tests of WLCG services via Nagios: done
JINR monitoring Network monitoring information system - more than 423 network nodes are in round-the-clock monitoring
Services and Readiness
CMS-specific activity • Currentlycommissioning Tier-1 resourceforCMS: • Local Tests of CMS VO-services and CMS SW • The PhEDExLoadTest(tests of data transfer links) • Job Robot Tests (or testsviaHammerCloud) • Long-running CPU intensive jobs • Long-running I/O intensive jobs • PHDEDX transferredofRAW inputdatatoourstorageelementwithtransferefficiencyaround 90% • Preparedservicesanddatastorageforthereprocessingof 20128TeVreprocessing
Data transfer link to CERN CMS Tier-1 in Dashborad
Frames for Grid cooperation of JINR • Worldwide LHC Computing Grid (WLCG) • Enabling Grids for E-sciencE (EGEE) - Now is EGI-InSPIRE • RDIG Development • CERN-RFBR project “GridMonitoringfrom VO perspective” • BMBF grant “Development of the grid-infrastructure and tools to provide joint investigations performed with participation of JINR and German research centers” • “Development of grid segment for the LHC experiments” was supported in frames of JINR-South Africa cooperation agreement; • Development of grid segment at Cairo University and its integration to the JINR GridEdu infrastructure • JINR - FZU AS Czech Republic Project “The grid for the physics experiments” • NASU-RFBR project “Development and support of LIT JINR and NSC KIPT grid-infrastructures for distributed CMS data processing of the LHC operation” • JINR-Romania cooperation Hulubei-Meshcheryakovprogramme • JINR-Moldova cooperation (MD-GRID, RENAM) • JINR-Mongolia cooperation (Mongol-Grid) 22
Staffing Korenkov V. Mitsyn V. Dolbilov A. Trofimov V. Shmatov S.
Milestones of the JINR CMS Tier-1 Deployment and Commissioning
Main tasks for next years • Engineering infrastructure (system of uninterrupted power supply and climate-control) • High-speed reliable network infrastructure with the allocated reserved channel to CERN (LHCOPN) • Computing system and storage system on the basis of disk arrays and tape libraries of high capacity • 100% reliability and availability.
US-BNL CERN Bologna/CNAF Ca-TRIUMF Russia: NRC KI Taipei/ASGC NDGF US-FNAL JINR UK-RAL Amsterdam/NIKHEF-SARA De-FZK Barcelona/PIC 26 June 2009 Lyon/CCIN2P3
The 6th International Conference "Distributed Computing and Grid-technologies in Science and Education" (GRID’2014) Dubna, 30 June-5 July 2014 GRID’2012 Conference 22 countries, 256 participants, 40 Universities and Institutes from Russia, 31 Plenary, 89 Section talks
Conclusions • In 2012-2013 CMS Tier1 prototype was created in Dubna • Disk & server installation and tests • Prototype tape system installation and tests • Organization of network infrastructure and connectivity to CERN via GEANT • Registration in GOC DB and APEL • Tests of WLCG services via Nagios • CMS-specific tests • Commissioning data transfer links (T0-T1, T1-T1, T1-T2) in progress • We expect to meet the start of next LHC run with full resources required (for the end of 2014)