Presentation is loading. Please wait.

Presentation is loading. Please wait.

Distributed Facilities for U.S. ATLAS Rob Gardner Indiana University PCAP Review of U.S. ATLAS Physics and Computing Project Argonne National Laboratory.

Similar presentations


Presentation on theme: "Distributed Facilities for U.S. ATLAS Rob Gardner Indiana University PCAP Review of U.S. ATLAS Physics and Computing Project Argonne National Laboratory."— Presentation transcript:

1 Distributed Facilities for U.S. ATLAS Rob Gardner Indiana University PCAP Review of U.S. ATLAS Physics and Computing Project Argonne National Laboratory OCTOBER 30, 2001

2 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 2 Outline  Requirements  Approach  Organization  Resource Requirements, current funding  Schedule  Grid Testbed and Prototype Tier 2 development  US LHC Common Computing Projects  Summary

3 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 3 Distributed IT Infrastructure  A wide area computational infrastructure for U.S. ATLAS  A network of distributed computing devices  A network of distributed data caches & stores  Connectivity  Physicists with data  Computers with data (at all scales)  Physicists with each other (collaboration)  Distributed information, portals  Efforts  Data Grid R&D (see talks by Wenaus, Schopf)  Prototype Tier 2 sites at Boston and Indiana  Networking and Grid Testbed coordinated by Ed May (ANL)

4 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 4 Requirements  Access  Efficient access to resources at the Tier 1 facility  Data distribution to remote computing devices  Information  A secure infrastructure to locate, monitor and manage collections of distributed resources  Analysis planning framework  Resource estimation  “Matchmaker” tools to optimally connect physicist+CPU+data+etc…  Scalable  Add arbitrary large numbers of computing devices as they become available  Add arbitrarily large numbers of data sources as they become available

5 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 5 Approach  ~5 strategic remote sites (Tier 2s)  Scale of each facility:  MONARC estimates  ATLAS NCB/WWC (World Wide Computing Group)  National Tier 1 facility  209K Spec95  365TB Online disk  2 PB tertiary  Tier 2 = Tier 1 * 20%  Networking

6 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 6 Organization  Facilities Subproject 2.3.2  Coordination:  Tier 2 centers (Gardner, Shank)  Testbed (May)  Networking (McKee)  Grid monitoring (Yu)

7 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 7 Role of Tier 2 Centers  User Analysis  Standard configuration optimized for analysis at the AOD level  ESD objects required for some analyses  Primary Resource for Monte Carlo Simulation  Data distribution caches (depends on distribution model)  Remote tertiary data stores  HSM services for quick, local AOD archival  MC data of all types (GEN, RAW, ESD, AOD, LOD) from all Tier 2’s & users  Relieve pressure, improve efficiency of Tier 1 systems  Effective use with grid software

8 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 8 Typical Tier 2  CPU: 50K SpecInt95 (t1: 209K SI95)  Commodity Pentium/Linux  Estimated 144 Dual Processor Nodes (t1: 640 nodes)  Online Storage: 70 TB Disk (t1: 365 TB)  High Performance Storage Area Network  Baseline: Fiber Channel Raid Array

9 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 9 ‘Remote’ Data Stores  Exploit existing infrastructure  mass store infrastructure at 2 of the 5 Tier 2 centers  Assume existing HPSS or equivalent license, tape silo, robot  Augment with drives, media, mover nodes, and disk cache  Each site contributes 0.3-0.5 PB store  AOD archival, MC ESD+AOD archival

10 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 10 Personnel (Lehman 11/00) MANPOWER ESTIMATE SUMMARY IN FTEs WBSNo:2Funding Type:Infrastructure11/13/00 8:08:38 PM Description:US ATLAS ComputingInstitutions:AllFunding Source :All FY 01FY 02FY 03FY 04FY 05FY 06Total IT I1.04.06.010.010.07.038.0 IT II.01.02.02.05.05.015.0 Physicist1.01.01.01.01.0.05.0 TOTAL LABOR 2.06.09.013.016.012.058.0 FTE’s NSF funded (GriPhyN and ½ iVDGL reorganized into Software): 1.0 1.0 1.01.0 1.0 0.0 5.0 GriPhyN 0.0 3.0 3.03.0 3.0 3.0 15.0 iVDGL 0.0 0.5 0.5 0.5 0.0 0.0 1.5 PPDG 0.0 0.8 1.0 1.0 0.0 0.0 2.8 ITR2 telemetry 0.0 0.5 1.0 1.0 1.0 1.0 4.5 iVDGL grid ops 1.0 5.8 6.5 6.5 5.0 4.0 28.8 total update10/01

11 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 11 Tier 2 Costs (Lehman 11/01)

12 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 12 Funding for Tier 2 Centers  Additional funding for prototype tier 2 centers and for permanent tier 2 centers will need to be found

13 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 13 Schedule  R&D Tier 2 centers  Two prototype tier 2 sites selected in 01: Boston U and Indiana U  Initial (university funded) centers established in 01  Support analysis of DC1 data in summer 02  DC2 production and analysis  Production Tier 2’s – FY ‘04 & FY ‘05  Operation – FY ‘05, FY ‘06 & beyond  Full Scale System Operation, 20% (‘05) to 100% (‘06) (as for Tier 1)

14 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 14 Persistent Grid Testbed for US-ATLAS ATLAS-US PCAP Meeting at ANL Oct 30, 2001 Ed May Argonne National Laboratory E. May

15 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 15 Background & Motivation Based on previous meetings of the US groups, in particular Summer 2000 at IU, organizational meeting Winter 2000-2001 at UM.Based on previous meetings of the US groups, in particular Summer 2000 at IU, organizational meeting Winter 2000-2001 at UM. Establish a persistent grid test-bed of ATLAS-US level 1, level 2 and other sites April 2001.Establish a persistent grid test-bed of ATLAS-US level 1, level 2 and other sites April 2001. Participating sites ANL, BNL, LBNL, BU, UM, IU, OU and UTA.Participating sites ANL, BNL, LBNL, BU, UM, IU, OU and UTA. Provide a focus for working with PPDG and GriPhyN. Ultimately with CERN & EDG.Provide a focus for working with PPDG and GriPhyN. Ultimately with CERN & EDG. E. May

16 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 16 Participants ANL HEPEd May, Jerry GieraltowskiANL HEPEd May, Jerry Gieraltowski LBNL(PDSF)Stu Loken, Shane CannonLBNL(PDSF)Stu Loken, Shane Cannon BNLRich Baker, Torre Wenus, Danton YuBNLRich Baker, Torre Wenus, Danton Yu Boston USaul Youssef, Jim ShankBoston USaul Youssef, Jim Shank Indiana URob GardnerIndiana URob Gardner Univ. of MichiganShawn Mckee, Erc MyersUniv. of MichiganShawn Mckee, Erc Myers Univ. of OklahomaHorst Severini, Pat SkubicUniv. of OklahomaHorst Severini, Pat Skubic Univ. of Texas at ArlingtonKaushik DeUniv. of Texas at ArlingtonKaushik De More information: http://www.usatlas.bnl.gov/computing/grid/http://www.usatlas.bnl.gov/computing/grid/ E. May

17 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 17 Calren Esnet, Abilene, Nton Esnet, Mren UC Berkeley LBNL-NERSC Esnet NPACI, Abilene Brookhaven National Laboratory Indiana University Boston University Argonne National Laboratory HPSS sites U Michigan University of Texas at Arlington 8 Sites in Testbed, ‘01 Oklahoma University Abilene

18 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 18 Planning & Implementation During 1 year (2001) implement testbed with globus 1.1.3 and 1.1.4During 1 year (2001) implement testbed with globus 1.1.3 and 1.1.4 Provide an environment for Grid developers and testers... relatively small number of friendly users. Not for production use.Provide an environment for Grid developers and testers... relatively small number of friendly users. Not for production use. Establishment of a technical working group with regular phone/VRVS meetings.Establishment of a technical working group with regular phone/VRVS meetings. production users.production users. E. May

19 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 19 P & I continued Long list of technical issues of what services and management issues.Long list of technical issues of what services and management issues. Scope of interest varies widely with institution: e.g.Scope of interest varies widely with institution: e.g. Data Cataloging and Replication (BNL) Objectivity Database issues (ANL) User & Account management (UM, IU) Remote job execution (BU) E. May

20 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 20 Activities & Accomplishments GridView grid testbed status (UTA)GridView grid testbed status (UTA) Magda distributed data manager prototype (BNL)Magda distributed data manager prototype (BNL) Pacman package manager (BU,BNL)Pacman package manager (BU,BNL) GRIPE A grid sign-up tool (IU)GRIPE A grid sign-up tool (IU) Distributed job management prototyping with Condor (BU,UTA,OUDistributed job management prototyping with Condor (BU,UTA,OU Testing of distributed data replication (MAGDA, GDMP, Globus) withTesting of distributed data replication (MAGDA, GDMP, Globus) with Atlas applications (Tilecal testbeam, ATLFast in Athena) (ANL,BU,OU)Atlas applications (Tilecal testbeam, ATLFast in Athena) (ANL,BU,OU) Network Performance and monitoring. (UM, BNL)Network Performance and monitoring. (UM, BNL) E. May

21 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 21 Testbed Software  Testbed has been functional for ~ 8 months  Accounts (individual user, group) created at all sites  Grid credentials (based on globus CA) distributed  To be updated with ESnet CA credentials  Grid software at each node in the site:  Globus 1.1.4  Condor 6.3  ATLAS core software distribution at some of the sites (for developers)  ATLAS related grid software: Pacman, Magda, Gridview  Start grid-wide applications in 02

22 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 22 Future Activities for Testbed Focus on Environments for ApplicationsFocus on Environments for Applications Compatibility with EDGCompatibility with EDG Preparations for Atlas Data Challenge 1 and 2Preparations for Atlas Data Challenge 1 and 2 E. May

23 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 23 IU Tier 2 Configuration  Gateway: atlas.uits.iupui.edu  Nodes: atlas01 – atlas016  400 MHz PII, 256MB  4.3 GB SCSI local disk  100 Mbs NIC  Switch  HP ProCurve 4000M 10/100Base-TX  Disk and Storage  /usr/lhc1 60 GB  /usr/lhc2 60 GB  lhc1.uits.iupui.edu 200 GB attached RAID, AFS  Generic atlas account into local HPSS

24 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 24 IU Notes  Tertiary storage  Currently IBM 3494 robot with ~10 TB ATLAS dedicated storage  New StorageTek to be installed in Feb 02 (capacity 360 TB)  HPSS software  Connectivity:  Located at Indianapolis campus (IUPUI) (OC12)  Better connectivity than Bloomington (DS3)  Bloomington-Indianapolis dark fiber project >1/2 complete, future installations could be located at IUB  Machine room adjacent to Abilene NOC, Global NOC  IU will develop grid operations center (iGOC) as part of iVDGL  Trouble-ticket system, monitoring, administrative

25 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 25 Boston University Tier 2 Linux Farm IBM R30 100 TB SGIs 230 cpu IBM sp 64 cpu RAID disk array NoX BU Router Campus Network Tufts Access GRID conference center High end graphics lab SGI Onyx x4 RE II, 9 O2 128 PIII Atlas dedicatedshared OC12 622 Mb/s internet 2 internet Harvard MIT

26 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 26 BU Notes  The 100 Terabyte mass store will be upgraded to 150 Terabytes.  Upgrading the local 100 Mb/s ethernet to Gigabit is being considered.  Nominal network bandwidth is proportional to the thickness of the purple lines.  For both IU and BU clusters, hardware funding in FY 02 will be used to optimize support for analysis of data challenge production  BU: large RAID purchase  IU: some RAID plus CPU upgrades

27 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 27 US LHC Common Computing Projects  Meeting of ATLAS & CMS PM’s, CERN 1/01  Identify possible common work areas (facilities, networking, database, grid)  Facilities Workshops (R Baker, V Odell):  BNL (3/01), FNAL (7/01), LBL (10/01)  Networking Workshops (S McKee):  IU (6/01), UM (10/01)

28 October 30, 2001 Rob Gardner Distributed Facilities for U.S. ATLAS 28 Summary  Prototype tier 2 centers chosen, U-funded resources in place  Persistent testbed for grid projects (PPDG, GriPhyN/iVDGL, EDG) established  US LHC coordination for facilities and networking established, working groups formed  Hiring for prototype tier 2 centers & ATLAS-grid integration begun  Facilities grid planning document; coherent with Software grid development; includes networking and facilities infrastructure


Download ppt "Distributed Facilities for U.S. ATLAS Rob Gardner Indiana University PCAP Review of U.S. ATLAS Physics and Computing Project Argonne National Laboratory."

Similar presentations


Ads by Google