Laboratory of Information Technologies. Distributed computing and Grid technologies. Grid-infrastructure infrastructure at JINR

Transcription

1 Laboratory of Information Technologies. Distributed computing and Grid technologies. Grid-infrastructure infrastructure at JINR Korenkov Vladimir (LIT,JINR) Dubna,,

2 Laboratory of Information Technologies

3 Specific task: Provision of theoretical and experimental studies conducted by the JINR Member State institutes at JINR and other scientific centres with modern telecommunication, network and information resources. In order to fulfill the task, it is necessary to provide: development of telecommunication channels of JINR with the JINR Member States on the basis of national and regional telecommunication networks; fault-tolerant tolerant operation and further development of the high- speed and protected local area network of JINR; development and maintenance of the distributed high- performance computing infrastructure and mass storage resources; information, algorithmic and software support of the research-and and-production activity of JINR; a reliable operation and development of the JINR Grid- segment as a component of the global Grid-infrastructure. infrastructure.

4 Cooperation Collaborators from the JINR Member States (protocols of cooperation with INRNE (Bulgaria), ArmeSFo (Armenia), FZK Karlsruhe GmbH (Germany), IHEPI TSU (Georgia), NC PHEP BSU (Belarus), KFTI NASU (Ukraine), etc.). Grants of Plenipotentiaries of the JINR Member States: Poland, Bulgaria, Slovakia, Romania. Agreement on Joint Research Programme Meshcheryakov-Hulubei LIT-Romania. BMBF grant JINR Network Security, Failover Measures, Network monitoring. CERN-JINR Cooperation Agreement on several topics, and JINR in the LCG Project (participation of JINR in the LCG project). JINR-South Africa Cooperation Agreement -project Development of Grid segment for the LHC experiments. Participation in common projects: Enabling Grids for E-sciencE (EGEE -a project co-funded by the European Commission through the Sixth Framework Programme, CERN-INTAS project and "SKIF-GRID project of the Program of the Belarusian-Russian Union State as a part of joint propositions of NASB and Federal Agency of Science and Innovations of Russian Federation The development and use of hard and software in GRID-technologies and advanced supercomputer systems SKIF in (SKIF-GRID), GridNNN (National Nanotechnological Network) Joint research grants from RFBR and Ministry of Science and Education with MSU and SINP MSU, Saratov University, IM SB RAS, IMM RAS, TU Gomel 4 (Belorussia).

5 Core e-science areas Numerical Analysis Visualization and Image Science Mathematical Modeling Distributed Resources Database Technology Parallel Algorithms and Performance Optimization Distributed Resources and Database Technology Numerical Analysis Visualization and Image Science Parallel Algorithms and Performance Optimization Mathematical Modeling 5

6 LIT JINR: two important projects completed Two large-scale very important projects have been completed in A grand presentation was held simultuneouslywith the meeting of the Programme Advisory Committee for Particle Physics on June 10. All organizations involved in the projects realisation together with the representatives of Ministry of Communications, Federal Agency for Science and Innovation attended the Presentation. Our community colleagues and prospective users - residents of the special economical zone - were also invited. 6

7 JINR - Moscow 20 Gbps telecommunication channel Collaboration with RSCC, RIPN, MSK-IX, JET Infosystems, Nortel

8 JINR Central Information and Computing Complex In 2009,, total CICC performance was 2400 ksi2k, the disk storage capacity 500 TB At present,, the CICC performance equals 2800 ksi2k and the disk storage capacity 1068TB (>1PB 1PB), 900TB for users from dcache and xtoord for data storage. Scheme of the CICC network connections

9 JINR Local Area Network Backbone (LAN) Plans: Step-by-step modernization of the JINR Backbone transfer to 10 Gbps Development and modernization of the control system of the JINR highway network Comprises 7250 computers and nodes, Users 4200, IP 8176, Remote VPN users (Lanpolis, Contact, TelecomMPK) ; High-speed transport (1Gbps/10 Gbps); Controlled-access; Partially isolated local traffic (8 divisions have own subnetworks with Cisco Catalyst 3550 as gateways); General network authorization system involves many services (AFS, batch, Grid, remote access, etc.) (AFS, batch, Grid, remote 9

10 Mathematical Support of JINR Experiments: mathematical support of particle and relativistic nuclear physics experiments performed in JINR and ones with JINR participation (LHC-CERN, GSI, RHIC, DESY, etc); design of mathematical methods and means of modeling physical processes and experimental data analysis; development and study of the new methods of 2D and 3D approximations and smoothing for increasing the efficiency of experimental data processing; development of software and computer complexes for experimental data processing; involvement of new workforce (young people) in data processing. 10

11 Numerical Investigation of Problems arising in the modeling of complex physical systems: creation of numerical algorithms and software for modeling complex physical systems; implementation of parallel algorithms proposed for calculations at multi-processor computer clusters using the MPI technology; development of new-generation computing techniques, including the gridification of the developed algorithms. 11

12 lit.jinr.ru 12

13 13

14 Challenges of scientific computing distributed computing and Grid 14

15 Industry Journey Old World Static Solo Physical Manual Application New World Dynamic Shared Virtual Automated Service 15

16 We do e-science e like in digital, distributed Science that is: Computationally intensive Operates on massive digital data sets Carried out in a distributed network environment High-Throughput vs High- Performance Computing: HTC: distributed (serial tasks), free cycles, cheap HPC: compact (parallel tasks), booked years ahead, expensive High-Energy Physics is a textbook example of e-science 16

17 New instruments, more data, more scientists, more computers WWW born in CERN 17

18 Why Grid? The Changing Nature of Work Collaborative & Dynamic Distributed & Heterogeneous Data & Computation Intensive Concurrent Innovation Cycles Project focused, globally distributed teams, spanning organizations within and beyond company boundaries Each team member/group brings own data, compute, & other resources into the project Access to computing and data resources must be coordinated across the collaboration Resources must be available to projects with strong QoS, & also reflect enterprise-wide biz priorities IT must adapt to this new reality 18

19 Five Emerging Models of Networked Computing From The Grid Distributed Computing synchronous processing High-Throughput Computing asynchronous processing On-Demand Computing dynamic resources Data-Intensive Computing databases Collaborative Computing scientists Ian Foster and Carl Kesselman, editors, The Grid: Blueprint for a New Computing Infrastructure, Morgan Kaufmann, 1999, 19

20 The Grid Enable coordinated resource sharing & problem solving in dynamic, multiinstitutional virtual organizations. (Source: The Anatomy of the Grid ) Access to shared resources Virtualization, allocation, management With predictable behaviors Provisioning, quality of service In dynamic, heterogeneous environments Standards-based interfaces and protocols 20

21 Grid computing Today there are many definitions of Grid computing: The definitive definition of a Grid is provided by Ian Foster in his article "What is the Grid? A Three Point Checklist" The three points of this checklist are: Computing resources are not administered centrally; Open standards are used; Non-trivial quality of service is achieved. 21

22 GÉANT2 Pan-European Backbone 34 NRENs, ~30M Users; 50k km Leased Lines 12k km Dark Fiber; Point to Point Services GN3 Next Gen. Network Projected Start Q Dark Fiber Core Among 19 Countries: u Austria u Belgium u Croatia u Czech Republic u Denmark u Finland u France u Germany u Hungary u Ireland u Italy u Netherlands u Norway u Slovakia u Slovenia u Spain u Sweden u Switzerland u United Kingdom 22

23 GEANT2 International Connectivity New Projects in u ALICE2 in and to Latin America u CAREN for Central Asia u FEAST: Feasibility Study for Sub-Saharan Africa Inter-Regional Connections u GEANT2/ESNet over US LHCNet for US Tier2/ EU Tier1 Connections u 4 10G Links to US R&E Nets u 1 GbE Connection to the Ubuntunet Alliance (Africa); also TENET in South Africa u ORIENT: CERNET + CSTNET (China) to Europe at 2.5G u TEIN2: 9 Asian Countries + Australia to Europe; TEIN3 Approved u EUMEDConnect: 11 Mediterranean + N. Africa u ALICE: 12 Latin American countries to GEANT2 at 622 Mbps u SuperSINET (Japan) at New York: 10G u India: 100 Mbps 23

24 Connectivity of JINR Member States INCREASE of the JINR - Moscow data link: 20 Gbps in Gbps in 2016 GÉANT3 Pan-European Backbone Consortium of 34 NRENs TEIN3: The Research and Education Network for Asia-Pacific Vietnam Bulgaria, Czech, Poland, Romania, Russia, Slovakia Ukraine, Belarus, Azerbaijan, Armenia, Georgia, Moldova,

25 Grid is a result of IT progress 25 Graph from The Triumph of the Light, G. Stix, Sci. T.Strizh Am. January (LIT, JINR) 2001

26 From the conventional HPC To the Grid 26

27 27

28 Job submission to the WLCG User Interface Replica Catalogue (RC) Information Service (IS) submitted waiting Resource Broker ready scheduled running Job Submission Service (JSS) Logging and Bookkeeping Computing Element (CE) Storage Element (SE) done outputready cleared 28

29 Global Community 29

30 Some history 1999 Monarc Project Early discussions on how to organise distributed computing for LHC EU DataGrid project middleware & testbed for an operational grid LHC Computing Grid LCG deploying the results of DataGrid to provide a production facility for LHC experiments EU EGEE project phase 1 starts from the LCG grid shared production infrastructure expanding to other communities and sciences EU EGEE-II Building on phase 1 Expanding applications and communities EU EGEE-III CERN 30

31 CERN - - European Organization for Nuclear Research Research/ Discovery Technology Training Collaborating 31

32 Large Hadron Collider Start-up of the Large Hadron Collider (LHC), one of the largest and truly global scientific projects ever, is the most exciting turning point in particle physics. CMS LHCb ALICE LHC ring: 27 km circumference ATLAS

33 ATLAS Number of scientists: 1800 Number of institutes: 164 Number of countries: 35 33

34 Tier 0 at CERN: Acquisition, First pass reconstruction, Storage & Distribution T.Strizh Ian.Bird@cern.ch (LIT, JINR) 1.25 GB/sec (ions) 34

35 Tier 0 Tier 1 Tier 2 Tier-0 (CERN): Data recording Initial data reconstruction Data distribution Tier-1 (11 centres): Permanent storage Re-processing Analysis Tier-2 (>200 centres): Simulation End-user analysis T.Strizh Ian.Bird@cern.ch (LIT, JINR) 35

36 DataGrid Architecture Local Computing Local Application Local Database Grid Grid Application Layer Job Management Data Management Metadata Management Object to File Mapping Collective Services Information & Monitoring Replica Manager Grid Scheduler Grid Underlying Grid Services Database Services Computing Element Services Storage Element Services Replica Catalog Authorization Authentication & Accounting Logging & Bookkeeping Fabric Fabric services Resource Management Configuration Management Monitoring and Fault Tolerance Node Installation & Management Fabric Storage Management 36

37 EGEE (Enabling Grids for E-sciencE) The aim of the project is to create a global Pan-European computing infrastructure of a Grid type. -Integrate regional Grid efforts -Represent leading grid activities in Europe 10 Federations, 27 Countries, 70 Organizations 37

38 350 sites 55 countries 150,000 CPUs 26 PetaBytes (Disk) 40 PetaBytes (Tape) >15,000 users >300 Vos 12 mln jobs/month Астрономия иастрофизика Безопасность населения Вычислительная химия Вычислительные науки/программирование Физика конденсированного состояния Науки оземле Синтез Физика высоких энергий Науки ожизни 38

39 39

40 glite Middleware BDII Information Services MON EGEE Maintained Components General Services Logging & Workload Book Management keeping Service Service File Transfer Service Compute Element CREAM glexec BLAH LCG-CE Worker Node User User Access External User Interface Components Interface LHC File Catalogue Hydra AMGA Storage Element Disk Pool Manager dcache Virtual Organisation Membership Service Proxy Server Security Services SCAS Authz. Service LCAS & LCMAPS Physical Resources Bob Jones - EGEE09 40

41 glite 41

42 Bioinformatics and Grid Many large clusters utilized for Services Sequence similarity (BLAST queues) Research Molecular modeling (folding, docking) Training of novel predictors Jobs are typically short (3 minutes) But plenty (all against all ) Considerable preparation for single job (couple of gigabytes of data to transfer) 42

43 Biomedical applications Biomedicine is also a pilot application area More than 20 applications deployed and being ported Three sub domains Medical image processing Biomedicine Drug discovery Use Grid as platform for collaboration (don t t need same massive processing power or storage as HEP) 43

44 Applications Example: WISDOM Grid-enabled drug discovery process for neglected diseases In silico docking compute probability that potential drugs dock with target protein To speed up and reduce cost to develop new drugs WISDOM (World-wide In Silico ( Malaria Docking On : Three large-scale deployments with more than 6 centuries of computations achieved in 190 days 3,5TB of data produced Up to 5000 computers in 50 countries Some promising in-vitro tests, with relevant biological results. Tuesday, July 20, 2010 EGEE Applications 44

45 Radio astronomy needs Grid, too Enormous datasets, massive computing, innovative instrumentation Dozens of new surveys launched recently Many (10 100) terabytes per survey High data rates researchers per survey International collaborations (almost always) Data is non-proprietary (usually) 45

46 Computational Chemistry GEMS (Grid Enabled Molecular Simulator) application Calculation and fitting of electronic energies of atomic and molecular aggregates (using high level ab initio methods) The use of statistical kinetics and dynamics to study chemical processes Virtual Monitors Angular distributions Vibrational distributions Rotational distributions Many body systems End-User applications Nanotubes Life sciences Statistical Thermodynamics Molecular Virtual Reality Angular distribution Rotational distribution Vibrational distribution Many body system Angular distribution 46

47 Fusion Large Nuclear Fusion installations E.g. International Thermonuclear Experimental Reactor (ITER) Distributed data storage and handling needed Computing power needed for Making decisions in real time Solving kinetic transport particle orbits Stellarator optimization magnetic field to contain the plasma 47

48 Earth Science Applications Community Many small groups that aggregate for projects (and separate afterwards) The Earth Complex system Independent domains with interfaces Solid Earth Ocean Atmosphere Physics, chemistry and/or biology Applications Earth observation by satellite Seismology Hydrology Climate Meteorology, Space Weather Geosciences Mars Atmosphere Pollution Database Collection 48

49 Earth Sciences: Earthquake analysis Seismic software application determines: Epicentre, magnitude, mechanism May make it possible to predict future earthquakes Assess potential impact on specific regions Analysis of Indonesian earthquake (28 March 2005) Data from French seismic sensor network GEOSCOPE transmitted to IPGP within 12 hours after the earthquake Solution found within 30 hours after earthquake occurred 10 times faster on the Grid than on local computers Results Not an aftershock of December 2004 earthquake Different location (different part of fault line further south) Different mechanism Rapid analysis of earthquakes is important for relief efforts 49

50 How to set up your own Grid 50 Slide from O.Smirnova (NDGF)

51 What next? 51

52 European e-infrastructuree Need to prepare permanent, common Grid infrastructure Ensure the long-term sustainability of the European e-infrastructure independent of short project funding cycles Coordinate the integration and interaction between National Grid Infrastructures (NGIs) Operate the European level of the production Grid infrastructure for a wide range of scientific disciplines to link NGIs The T.Strizh EGEE (LIT, project JINR) -Bob Jones -EGEE'08-22 September

53 Grid establishes itself as a scientific tool EU invests into Grids: Over 450 m invested in last 8 years to support Grid research and e-infrastructure e deployments One Call towards the end of 2009 (indicative budget 95 m): m Create the European Grid Initiative (EGI) Software and services in support of EGI and HPC User Communities focus on e-infrastructure e needs of important user-communities Support Measures including international cooperation and prospective and strategic studies 53

54 What future holds ARC consortium (NorduGrid( NorduGrid,, NDGF, KnowARC et al), together with glite and UNICORE, contribute to creation of the Universal Middleware Distribution (UMD) for the European Grid Initiative (EGI) Sites and VOs that use ARC will get an access to the European e-science e infrastructure, just like those that use glite or UNICORE What about Clouds? Technically, very similar to Grids: distributed, service- oriented However, Clouds business model is closer to that of HPC Single administrative domain, carefully selected resources 54

55 The Future of Grids From e-infrastructures e to Knowledge Infrastructures Network infrastructure connects computing and data resources and allows their seamless usage via Grid infrastructures Federated resources and new technologies enable new application fields: Distributed digital libraries Distributed data mining Digital preservation of cultural heritage Data curation Knowledge Infrastructure Major Opportunity for Academic and Businesses alike 55 KNOWLEDGE. INFRASTRUCTURE GRID. INFRASTRUCTURE NETWORK. INFRASTRUCTURE 55

56 Mirco Mazzucato DUBNA Grids, clouds, supercomputers.. Grids, clouds, supercomputers, etc. Grids Supercomputers Collaborative environment Expensive Distributed resources Low latency interconnects (political/sociological) Applications peer reviewed Commodity hardware (also Parallel/coupled applications supercomputers) Traditional interfaces (login) (HEP) data management Also SC grids (DEISA, Teragrid) Complex interfaces (bug not feature) Many different problems: Amenable to different Clouds solutions Volunteer computing Proprietary (implementation) Simple mechanism to access Economies of scale in management No right answer millions CPUs Commodity hardware Difficult if (much) data involved Virtualisation for service provision and Control of environment Ł check encapsulating application environment Community building people Details of physical resources hidden Simple interfaces (too simple?) involved in Science Potential for huge amounts of real work 56 Ian Bird 56

57 Software-as as-a-service (SaaS( SaaS) Platform-as as-a-service (PaaS( PaaS) Infrastructure-as as-a-service (IaaS( IaaS) Everything as a Service (XaaS( XaaS)

58 MACS Lab DNA-Array Application Layer Radiology Application Grid Layer Virtual Laboratory layer.

59 Some Desktop Grids World Community Grid - IBM ( PCs Leiden Classical Grid - Education on Grid ( boinc.gorlaeus.net/) PCs SZTAKI - Hungarian initiative ( PCs AlmereGrid ( almeregrid.nl) ) PCs PS3GRID (Based on Playstations) (

60 Real World Problems Taking Us BEYOND PETASCALE 1 ZFlops 100 EFlops 10 EFlops 1 EFlops 100 PFlops 10 PFlops 1 PFlops 100 TFlops 10 TFlops 1 TFlops 100 GFlops 10 GFlops 1 GFlops What we can just model today with <100TF SUM Of Top500 #1 Example Aerodynamic Real Analysis: World Challenges: 1 Petaflops Full Laser modeling Optics: of an aircraft in all 10 conditions Petaflops Green Molecular airplanes Dynamics in Biology: 20 Petaflops Genetically Aerodynamic Design: tailored medicine 1 Exaflops Computational Cosmology: 10 Exaflops Understand the origin of the universe Turbulence in Physics: 100 Exaflops Synthetic Computational fuels Chemistry: everywhere 1 Zettaflops Source: Dr. Steve Chen, The Growing HPC Momentum in China, Accurate extreme June 30 weather prediction th, 2006, Dresden, Germany 100 MFlops

61 Reach Exascale by 2018 From GigFlops to ExaFlops ~1987 Sustained GigaFlop ~1997 Sustained TeraFlop 2008 Sustained PetaFlop ~2018 Sustained ExaFlop Note: Numbers are based on Linpack Benchmark. Dates are approximate. The pursuit of each milestone has led to important breakthroughs in science and engineering. Source: IDC In Pursuit of Petascale Computing: Initiatives Around the World, 2007

62

63 Суперкомпьютер МГУ Ломоносов

64 Japan Courtesy of Satoshi Matsuoka, Tokyo Institute of Technology, Japan, ISC-2010

65 Grid-infrastructure infrastructure at JINR

66 Development of the JINR Grid - environment 66

67 JINR in the Russian Data Intensive Grid infrastructure (RDIG) The Russian consortium RDIG (Russian Data Intensive Grid), was set up in September 2003 as a national federation in the EGEE project. Now the RDIG infrastructure comprises 17 Resource Centers with > 5000 CPU (12000 ksi2k) and > 3200 TB of disc storage. RDIG Resource Centres: ITEP JINR-LCG2 Kharkov-KIPT RRC-KI RU-Moscow-KIAM RU-Phys-SPbSU RU-Protvino-IHEP RU-SPbSU Ru-Troitsk-INR ru-impb-lcg2 ru-moscow-fian ru-moscow-gcras ru-moscow-mephi ru-pnpi-lcg2 ru-moscow-sinp - BY-NCPHEP 67

68 Development and maintenance of RDIG e-infrastructure - support of basic grid-services; - Support of Regional Operations Center (ROC); - Support of Resource Centers (RC) in Russia; - RDIG Certification Authority; - RDIG Monitoring and Accounting; - participation in integration, testing, certification of grid-software; - support of Users, Virtual Organization (VO) and application; - User & Administrator training and education; - Dissemination, outreach and Communication grid activities.

69 VOs Infrastructure VO's (all RC's): dteam ops Most RC support the WLCG/EGEE VO's Alice Atlas CMS LHCb Supported by some RC's: gear Biomed Fusion Regional VO's Ams, eearth, photon, rdteam, rgstest Flagship applications: LHC, Fusion (toward to ITER), nanotechnology Current interests from: medicine, engineering 69

70 LHC Computing Grid Project (LCG) The protocol between CERN, Russia and JINR on a participation in LCG Project has been approved in The tasks of the Russian institutes in the LCG have been defined as: LCG software testing; evaluation of new Grid technologies (e.g. Globus toolkit 3) in a context of using in the LCG; event generators repository, data base of physical events: support and development.

71 Worldwide LHC Computing Grid Project (WLCG) The tasks of the Russian institutes & JINR in the LCG (2009 years): Task 1. MW (glite) Testsuit (supervisor O. Keeble) Task 2. LCG vs Experiments (supervisor I. Bird) Task 3. LCG monitoring (supervisor J. Andreeva) Task 4/5. Genser/ MCDB ( supervisor A. Ribon)

72 Worldwide LHC Computing Grid Project (WLCG) The protocol between CERN, Russia and JINR on participation in LCG Project was approved in MoU on Worldwide LHC Computing Grid (WLCG) signed by JINR in October, 2007 The tasks of the JINR in the WLCG: WLCG-infrastructure support and development at JINR; participation in WLCG middleware testing/evaluation, participation in Service/Data Challenges, grid monitoring and accounting tools development; FTS-monitoring and testing; participation in ARDA activities in coordination with experiments; JINR LCG portal support and development; HEP applications; MCDB development; support of JINR Member States in the WLCG activities; User & Administrator training and education. 72

73 Integration, testing, certification of new MW Testing glite components: - Metadata catalog, Fireman catalog, gridftp, glite-amga metadata service, dcache, -testing glite for ATLAS and CMS, -- Collaboration with the condor-g team and development of monitoring for CONDOR -- development dashboard Evaluation of new MW: JINR, KIAM, SINP MSU -OMII (Open Middleware Infrastructure Institute) - Globus Toolkit 3 & 4;

74 Participation in WLCG MCDB Development MCDB access libraries have been integrated to CMSSW software Several improvements have been made in MCDB software New improved XML schema has been developed for High Energy Physics Markup Language (HepML( HepML) The program libraries have been developed to work with new HepML schema

75 HepWeb Overview Provides: WEB access to computing resources of LIT for Monte Carlo simulations of hadron-hadron, hadronnucleus, and nucleus-nucleus interactions, by means of most popular generators. Realization: service - oriented architecture. Goals: Monte Carlo simulations at the server Provide physicists with new calculation/simulation tools Mirror site of GENSER of the LHC Computing GRID project Provide physicists with informational and mathematical support Introduce young physicists into HEP world 75

76 RDIG monitoring&accounting Monitoring allows to keep an eye on parameters of Grid sites' operation in real time Accounting - resources utilization on Grid sites by virtual organizations and single users Monitored values CPUs - total /working / down/ free / busy Jobs - running / waiting Storage space - used / available Network - Available bandwidth Accounting values Number of submitted jobs Used CPU time Totally sum in seconds Normalized (with WNs productivity) Average time per job Waiting time Totally sum in seconds Average ratio waiting/used CPU time per job Physical memory Average per job JINR CICC 76

77 FTS (FILE TRANSFER SERVICE) MONITORING FOR WORLDWIDE LHC COMPUTING GRID (WLCG) PROJECT A monitoring systemis developed which provides a convenient and reliable tool for receiving detailed information about the FTS current stateand the analysis of errorson data transfer channels, maintaining FTS functionality and optimization of the technical support process. The system could seriously improve the FTS reliability and performance. 77

78 USER- INTERFACE AND VISUALIZATION SERVICE DEVELOPMENT FOR VIRTUAL ORGANIZATION SUPPORT IN HIGH ENERGY PHYSICS S. Mitsyn (LIT) LHC Project Support Grid Monitoring:Deals with decentralized structures involving a large amount of data. Its proper representation is an essential part of the monitoring process. Google Earth offers a quite informative and visually attractive representation which mapping Grid infrastructure objects, processes and events on a geographic map. 78

79 Portal egee-rdig.ruru 79

80 Deletion service for ATLAS Distributed Data Management (DDM) The ATLAS Distributed Data Management project (DQ2) is responsible for the replication, access and bookkeeping of ATLAS data across more than 100 distributed grid sites. One of the most important service of DQ2 is deletion service. This distributed service interacts with 3rd party grid middleware and the DQ2 catalogs to serve data deletion requests on the grid. Furthermore, it also takes care of retrial strategies, check-pointing transactions, and performance throttling, ensuring DQ2's scalability and fault tolerance. Works on Deletion service include: support current version of software. Step by step development of new version of deletion service (more stable, with extended monitoring system). Works in development included: building new interfaces between parts of deletion service (based on web service technology), creating new database schema, rebuilding deletion service core part, builds extended interfaces with mass storage systems, extend deletion monitoring system

81 Deletion service for ATLAS DDM. Current progress Created HTTP client-server components for deletion service (in preproduction testing) Created monitoring web site backend with using AJAX technology and Django framework (in preproduction testing) with data export in JSON format Created new database schema (including triggers, sequences, etc) In works migration scenario for new DB schema, extension monitoring system. In nearest future: rebuilding deletion service core Finish of main development works for new version of deletion service (in current roadmap) excepted at the and of summer.

82 ATLAS Tier-3 status in Dubna We plan to try PROOF/Xrootd to run ATLAS analysis. LNP analysis cluster LIT analysis facility Central JINR SE is used as data storage

83 Remote ATLAS Control Room in Dubna MOTIVATION Monitoring of the detector at any time Participation of the subsystem experts from Dubna in the shifts and data quality checks remotely Training the shifters before they come to CERN

84 Production Normalised CPU time per EGEE Region (June 2009-May 2010) 84

85 Russia and JINR Normalized CPU time per SITE (June May 2010) 85

86 Production Normalised CPU time per EGEE site for VO LHC (June 2009 May 2010) GRID-site CPU time Num CPU (Core) FZK-LCG 41,936, RAL-LCG2 27,043, IN2P3-CC 24,462, GRIF 20,083, NIKHEF-ELPROD 17,162, DESY-HH 16,413, UKI-GLASGOW 16,404, INFN-T1 15,683, PIC 14,520, SARA-MATRIX 14,492, TRIUMF-LCG2 14,490, IN2P3-CC-T2 14,059, JINR 13,503, NDGF-T1 12,519,

87 Development of the JINR Grid-environment Network level: links between Moscow and Dubna on the basis of state-of-the-art technologies DWDM and 10Gb Ethernet. JINR Local area network : JINR High-speed backbone construction 10Gbps Resource level: requirements of the LHC experiments stimulate the development of a global Grid-infrastructure, together with the resource centers of all the cooperating organizations. First of all, this is of primary concern for such large research centers as the JINR. To reach effective processing and analysis of the experimental data, further increase in the JINR CICC performance and disk space is needed. 87

88 Grid training and education distributed training infrastructure: glite user trainings for students of Dubna University and University Centre of JINR, grid site administrators trainings for JINR member-states, testbed for grid developers, testbed for middleware evaluation, GILDA cooperation 88

89 Participation in GridNNN project Grid support for Russian national nanotechnology network To provide for science and industry an effective access to the distributed computational, informational and networking facilities Expecting breakthrough in nanotechnologies Supported by the special federal program Main points based on a network of supercomputers (about 15-30) has two grid operations centers (main and backup) is a set of grid services with unified interface partially based on Globus Toolkit 4 89

90 GridNNN infrastructure 10 resource centers at the moment in different regions of Russia RRC KI, «Chebyshev» (MSU), IPCP RAS, CC FEB RAS, ICMM RAS, JINR, SINP MSU, PNPI, KNC RAS, SPbSU 90

91 JINR in GridNNN projects Development activities Monitoring and accounting system Registration system of grid services and sites User support service Application area Virtual organization for molecular dynamics calculations Adjustment of some JINR's applications to parrallel execution in GridNNN environment Infrastructure for applications development and testing 91

92 Grid cooperation with JINR member-states Common project, grants (Czech, Slovak, Germany, South Africa, Belarus, Bulgaria, Ukraine, Romania Protocols, agreements (Armenia, Belarus, Bulgaria, Moldova, Poland, Czech, Slovak grid site administrators trainings (Ukraine, Romania, Uzbekistan, Azerbaijan Courses, Lectures, Practical training for students and users (Egipt, Bulgaria Consulting (Cuba, Georgia, Kazakhstan, Mongolia, Vietnam, Korea 92

93 Frames for Grid cooperation of JINR Worldwide LHC Computing Grid (WLCG); Enabling Grids for E-sciencE (EGEE); RDIG Development (Project of FASI) CERN-RFBR project Grid Monitoring from 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; NATO project "DREAMS-ASIA (Development of grid EnAbling technology in Medicine&Science for Central ASIA); JINR -FZU AS Czech Republic Project The GRID infrastructure for the physics experiments NASU-RFBR project Development and support of LIT JINR and NSC KIPT grid-infrastructures for distributed CMS (CERN) data processing during the first two years of the Large Hadron Collider operation Project Elaboration of distributed computing JINR-Armenia grid-infrastructure for carrying out mutual scientific investigations JINR-Romania cooperation Hulubei-Meshcheryakov programme Project "SKIF-GRID" (Program of Belarussian-Russian Union State). Project GridNNN (National Nanotechnological Net) 93

94 Applied level developments of the JINR Grid-environment (I) 1. The applied level covers the user applications working in a virtual organization (VO) environment which comprises both users and owners of computing resources. 2. In the existing Grid-systems, a VO defines a collaboration of specialists in some area, who combine their efforts to achieve a common aim. 3. The virtual organization is a flexible structure that can be formed dynamically and may have a limited life-time. VOs working within the WLCG project are the VOs on the LHC experiments - ATLAS, CMS, Alice, LHCb,, the first three being carried out with the noticeable and direct participation of the JINR. Nowadays, as a Grid-segment of the EGEE/RDIG, the JINR CICC supports computations of the virtual organizations registered in RDIG ( LHC experiments, BioMed,, PHOTON, eearth,, Fusion, HONE, Panda. In the future, as the interest arises at a large-scale level, VOs can be organized at JINR in the fields of nuclear physics and condensed matter physics and, most probably, in the new promising direction related to the research of the nanostructure properties. 94

95 Applied level developments of the JINR Grid-environment (II) The creation of new VOs gets possible and necessary under maturation of the algorithmic approaches to the problem solution, the development of corresponding mathematical methods and tools: methods and tools for simulation of physical processes and analysis of experimental data software and computer complexes for experimental data processing; numerical methods, algorithms and software for modeling complex physical systems; methods, algorithms and software of computer algebra; new computing paradigms; adaptation of specialized software for solving problems within the Grid- environment. 95

96 4-rd International Conference "Distributed Computing and Grid-technologies in Science and Education 28 June 3 July,

97 WEB-PORTAL GRID AT JINR ГРИДВОИЯИ : A new informational resource has been created at JINR: web-portal GRID AT JINR. The content includes the detailed information on the JINR grid-site and JINR s participation in grid projects: КОНЦЕПЦИЯГРИД ГРИД-технологии ГРИД-проекты Консорциум РДИГ ГРИД-САЙТ ОИЯИ Инфраструктура исервисы Схема Статистика Поддержка ВОиэкспериментов ATLAS CMS CBM и PANDA HONE Какстатьпользователем ОИЯИВГРИД-ПРОЕКТАХ WLCG ГридННС EGEE Проекты РФФИ ПроектыИНТАС СКИФ-ГРИД ТЕСТИРОВАНИЕГРИД-ПО СТРАНЫ-УЧАСТНИЦЫОИЯИ МОНИТОРИНГИАККАУНТИНГ RDIG-мониторинг dcache-мониторинг Dashboard FTS-мониторинг Н1 МС-мониторинг ГРИД-КОНФЕРЕНЦИИ GRID NEC ОБУЧЕНИЕ Учебнаягрид-инфраструктура Курсыилекции Учебныематериалы ДОКУМЕНТАЦИЯ Статьи Учебныематериалы НОВОСТИ КОНТАКТЫ

98 WEB-PORTAL GRID AT JINR ГРИДВОИЯИ :

99 Useful References: Grid Café: OPEN GRID FORUM: GLOBUS: TERAGRID RID: Open Science Grid: opensciencegrid.org/ LCG: lcg.web.cern.ch/lcg/ EGEE: EGEE-RDIG: egee-rdig.ru EGI: egi.eu/ International Science Grid this Week:

100 The blind men and the elephant in the room Cyberinfrastructure SaaS SOA Shared Infrastructure/ Shared Services Web 2.0 Grids Automation Virtualization

101 Thank you for your attention! 101