U.S. ATLAS Computing Facilities (Overview) Bruce G. Gibbard Brookhaven National Laboratory Review of U.S. LHC Software and Computing Projects Fermi National.

Slides:



Advertisements
Similar presentations
31/03/00 CMS(UK)Glenn Patrick What is the CMS(UK) Data Model? Assume that CMS software is available at every UK institute connected by some infrastructure.
Advertisements

Resources for the ATLAS Offline Computing Basis for the Estimates ATLAS Distributed Computing Model Cost Estimates Present Status Sharing of Resources.
Distributed IT Infrastructure for U.S. ATLAS Rob Gardner Indiana University DOE/NSF Review of U.S. ATLAS and CMS Computing Projects Brookhaven National.
1 Software & Grid Middleware for Tier 2 Centers Rob Gardner Indiana University DOE/NSF Review of U.S. ATLAS and CMS Computing Projects Brookhaven National.
Current Monte Carlo calculation activities in ATLAS (ATLAS Data Challenges) Oxana Smirnova LCG/ATLAS, Lund University SWEGRID Seminar (April 9, 2003, Uppsala)
Title US-CMS User Facilities Vivian O’Dell US CMS Physics Meeting May 18, 2001.
US ATLAS Distributed IT Infrastructure Rob Gardner Indiana University October 26, 2000
DATA PRESERVATION IN ALICE FEDERICO CARMINATI. MOTIVATION ALICE is a 150 M CHF investment by a large scientific community The ALICE data is unique and.
CERN/IT/DB Multi-PB Distributed Databases Jamie Shiers IT Division, DB Group, CERN, Geneva, Switzerland February 2001.
U.S. ATLAS Physics and Computing Budget and Schedule Review John Huth Harvard University DOE/NSF Review of U.S. ATLAS and CMS Computing Projects Brookhaven.
GridPP Steve Lloyd, Chair of the GridPP Collaboration Board.
October 24, 2000Milestones, Funding of USCMS S&C Matthias Kasemann1 US CMS Software and Computing Milestones and Funding Profiles Matthias Kasemann Fermilab.
LCG Milestones for Deployment, Fabric, & Grid Technology Ian Bird LCG Deployment Area Manager PEB 3-Dec-2002.
CMS Report – GridPP Collaboration Meeting VI Peter Hobson, Brunel University30/1/2003 CMS Status and Plans Progress towards GridPP milestones Workload.
High Energy Physics At OSCER A User Perspective OU Supercomputing Symposium 2003 Joel Snow, Langston U.
November 15, 2000US CMS Tier2 Plans Matthias Kasemann1 US CMS Software and Computing Tier 2 Center Plans Matthias Kasemann Fermilab DOE/NSF Baseline Review.
9/16/2000Ian Bird/JLAB1 Planning for JLAB Computational Resources Ian Bird.
Computing for HEP in the Czech Republic Jiří Chudoba Institute of Physics, AS CR, Prague.
Distributed Facilities for U.S. ATLAS Rob Gardner Indiana University PCAP Review of U.S. ATLAS Physics and Computing Project Argonne National Laboratory.
D0 SAM – status and needs Plagarized from: D0 Experiment SAM Project Fermilab Computing Division.
Data GRID Activity in Japan Yoshiyuki WATASE KEK (High energy Accelerator Research Organization) Tsukuba, Japan
U.T. Arlington High Energy Physics Research Summary of Activities August 1, 2001.
U.S. ATLAS Computing Facilities Bruce G. Gibbard Brookhaven National Laboratory Review of U.S. LHC Software and Computing Projects LBNL, Berkeley, California.
Fermilab User Facility US-CMS User Facility and Regional Center at Fermilab Matthias Kasemann FNAL.
LHC Computing Review - Resources ATLAS Resource Issues John Huth Harvard University.
Tier 1 Facility Status and Current Activities Rich Baker Brookhaven National Laboratory NSF/DOE Review of ATLAS Computing June 20, 2002.
Finnish DataGrid meeting, CSC, Otaniemi, V. Karimäki (HIP) DataGrid meeting, CSC V. Karimäki (HIP) V. Karimäki (HIP) Otaniemi, 28 August, 2000.
CRISP & SKA WP19 Status. Overview Staffing SKA Preconstruction phase Tiered Data Delivery Infrastructure Prototype deployment.
14 Aug 08DOE Review John Huth ATLAS Computing at Harvard John Huth.
7April 2000F Harris LHCb Software Workshop 1 LHCb planning on EU GRID activities (for discussion) F Harris.
June 02 John Huth, LHC Computing 1 U.S. ATLAS Overview  Project ManagementJ. Huth  SoftwareT.Wenaus  ArchitectureD. Quarrie  PhysicsI. Hinchliffe 
Questions for ATLAS  How can the US ATLAS costs per SW FTE be lowered?  Is the scope of the T1 facility matched to the foreseen physics requirements.
U.S. ATLAS Tier 1 Planning Rich Baker Brookhaven National Laboratory US ATLAS Computing Advisory Panel Meeting Argonne National Laboratory October 30-31,
O AK R IDGE N ATIONAL L ABORATORY U.S. D EPARTMENT OF E NERGY Facilities and How They Are Used ORNL/Probe Randy Burris Dan Million – facility administrator.
Atlas CAP Closeout Thanks to all the presenters for excellent and frank presentations Thanks to all the presenters for excellent and frank presentations.
U.S. ATLAS Project Overview John Huth Harvard University LHC Computing Review FNAL November 2001.
BNL Wide Area Data Transfer for RHIC & ATLAS: Experience and Plans Bruce G. Gibbard CHEP 2006 Mumbai, India.
US ATLAS Tier 1 Facility Rich Baker Brookhaven National Laboratory DOE/NSF Review of U.S. ATLAS and CMS Computing Projects Brookhaven National Laboratory.
BNL Tier 1 Service Planning & Monitoring Bruce G. Gibbard GDB 5-6 August 2006.
US ATLAS Tier 1 Facility Rich Baker Brookhaven National Laboratory Review of U.S. LHC Software and Computing Projects Fermi National Laboratory November.
Networking Shawn McKee University of Michigan DOE/NSF Review November 29, 2001.
Grid User Interface for ATLAS & LHCb A more recent UK mini production used input data stored on RAL’s tape server, the requirements in JDL and the IC Resource.
ATLAS Tier 1 at BNL Overview Bruce G. Gibbard Grid Deployment Board BNL 5-6 September 2006.
6/23/2005 R. GARDNER OSG Baseline Services 1 OSG Baseline Services In my talk I’d like to discuss two questions:  What capabilities are we aiming for.
ATLAS WAN Requirements at BNL Slides Extracted From Presentation Given By Bruce G. Gibbard 13 December 2004.
U.S. ATLAS Computing Facilities Bruce G. Gibbard GDB Meeting 16 March 2005.
Networking Shawn McKee University of Michigan PCAP Review October 30, 2001.
The following is a collection of slides from a few recent talks on computing for ATLAS in Canada, plus a few new ones. I might refer to all of them, I.
ATLAS Midwest Tier2 University of Chicago Indiana University Rob Gardner Computation and Enrico Fermi Institutes University of Chicago WLCG Collaboration.
U.S. ATLAS Computing Facilities (Overview) Bruce G. Gibbard Brookhaven National Laboratory US ATLAS Computing Advisory Panel Meeting Argonne National Laboratory.
U.S. ATLAS Computing Facilities Overview Bruce G. Gibbard Brookhaven National Laboratory U.S. LHC Software and Computing Review Brookhaven National Laboratory.
Tier 1 at Brookhaven (US / ATLAS) Bruce G. Gibbard LCG Workshop CERN March 2004.
The ATLAS Computing Model and USATLAS Tier-2/Tier-3 Meeting Shawn McKee University of Michigan Joint Techs, FNAL July 16 th, 2007.
ATLAS Computing Requirements LHCC - 19 March ATLAS Computing Requirements for 2007 and beyond.
Computing Issues for the ATLAS SWT2. What is SWT2? SWT2 is the U.S. ATLAS Southwestern Tier 2 Consortium UTA is lead institution, along with University.
U.S. ATLAS Computing Facilities Bruce G. Gibbard Brookhaven National Laboratory Mid-year Review of U.S. LHC Software and Computing Projects NSF Headquarters,
US ATLAS Tier 1 Facility Rich Baker Deputy Director US ATLAS Computing Facilities October 26, 2000.
U.S. ATLAS Computing Facilities DOE/NFS Review of US LHC Software & Computing Projects Bruce G. Gibbard, BNL January 2000.
U.S. ATLAS Computing Facilities U.S. ATLAS Physics & Computing Review Bruce G. Gibbard, BNL January 2000.
LHCb Current Understanding of Italian Tier-n Centres Domenico Galli, Umberto Marconi Roma, January 23, 2001.
WLCG Status Report Ian Bird Austrian Tier 2 Workshop 22 nd June, 2010.
Meeting with University of Malta| CERN, May 18, 2015 | Predrag Buncic ALICE Computing in Run 2+ P. Buncic 1.
10-Feb-00 CERN HepCCC Grid Initiative ATLAS meeting – 16 February 2000 Les Robertson CERN/IT.
Bernd Panzer-Steindel CERN/IT/ADC1 Medium Term Issues for the Data Challenges.
Hall D Computing Facilities Ian Bird 16 March 2001.
LQCD Computing Project Overview
Readiness of ATLAS Computing - A personal view
Dagmar Adamova (NPI AS CR Prague/Rez) and Maarten Litmaath (CERN)
Nuclear Physics Data Management Needs Bruce G. Gibbard
LHCb thinking on Regional Centres and Related activities (GRIDs)
Presentation transcript:

U.S. ATLAS Computing Facilities (Overview) Bruce G. Gibbard Brookhaven National Laboratory Review of U.S. LHC Software and Computing Projects Fermi National Laboratory November 27-30, 2001

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 2 Outline  US ATLAS Computing Facilities Definition  Mission  Architecture & Elements  Motivation for Revision of the Computing Facilities Plan  Schedule  Computing Model & Associated Requirements  Technology Evolution  Tier 1 Budgetary Guidance  Tier 1 Personnel, Capacity, & Cost Profiles for New Facilities Plan

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 3 US ATLAS Computing Facilities Mission  Facilities procured, installed and operated  …to meet U.S. “MOU” obligations to ATLAS  Direct IT support (Monte Carlo generation, for example)  Support for detector construction, testing, and calibration  Support for software development and testing  …to enable effective participation by US physicists in the ATLAS physics program !  Direct access to and analysis of physics data sets  Simulation, re-reconstruction, and reorganization of data as required to complete such analyses

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 4 Element of US ATLAS Computing Facilities  A Hierarchy of Grid Connected Distributed Resources Including:  Tier 1 Facility Located at Brookhaven – Rich Baker / Bruce Gibbard  Operational at < 0.5% level  5 Permanent Tier 2 Facilities ( to be Selected in April ’03 )  2 Prototype Tier 2’s selected earlier this year and now active  Indiana University – Rob Gardner  Boston University – Jim Shank  Tier 3 / Institutional Facilities  Several currently active; most candidate to become Tier 2’s  Univ. of California at Berkeley, Univ. of Michigan, Univ. of Oklahoma, Univ. of Texas at Arlington, Argonne Nat. Lab.  Distribute IT Infrastructure – Rob Gardner  US ATLAS Persistent Grid Testbed – Ed May  HEP Networking – Shawn McKee  Coupled to Grid Projects with designated liaisons  PPDG – Torre Wenaus  GriPhyN – Rob Gardner  iVDGL – Rob Gardner  EU Data Grid – Craig Tull

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 5 Tier 2’s  Mission of Tier 2’s for US ATLAS  A primary resource for simulation  Empower individual institutions and small groups to do relatively autonomous analysis using high performance regional networks and more directly accessible and locally managed resources  Prototype Tier 2’s were selected based on their ability to contribute rapidly to Grid architecture development  Goal in future Tier 2 selections will be to leverage particularly strong institutional resources of value to ATLAS  Aggregate of the 5 Tier 2’s is expected to be comparable to Tier 1 in CPU and disk capacity available for analysis

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 6 US ATLAS Persistent Grid Testbed Calren Esnet, Abilene, Nton Esnet, Mren UC Berkeley LBNL-NERSC Esne t NPACI, Abilene Brookhaven National Laboratory Indiana University Boston University Argonne National Laboratory U Michigan Oklahoma University Abilene Prototype Tier 2s HPSS sites University of Texas At Arlington

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 7 Evolution of US ATLAS Facilities Plan  In Responds to Changes or Potential Changes in  Schedule  Computing Model & Requirements  Technology  Budgetary Guidance

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 8 Changes in Schedule  LHC start-up projected to be a year later  2005/2006  2006/2007  ATLAS Data Challenges (DC’s) have, so far, stayed fixed  DC0 – Nov/Dec 2001 – 10 5 events  Software continuity test  DC1 – Feb/Jul 2002 – 10 7 events  ~1% scale test  DC2 – Jan/Sep 2003 – 10 8 events  ~10% scale test  A serious functionality & capacity exercise  A high level of US ATLAS facilities participation is deemed very important

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 9 Computing Model and Requirements  Nominal model was:  At Tier 0 (CERN)  Raw  ESD/AOD/TAG pass done, result shipped to Tier 1’s  At Tier 1’s (six anticipated for ATLAS)  TAG/AOD/~25% of ESD on Disk, Tertiary storage for remainder of ESD  Selection passes through complete ESD ~monthly  Analysis of TAG/AOD/selected ESD/etc. (n-tuples) on disk for analysis pass by ~200 users within 4 hours  At Tier 2’s (five in U.S.)  Data access primarily via Tier 1 (to control load on CERN and transatlantic link)  Support ~50 users as above but frequent access ESD on disk at Tier 1 likely  Serious limitations are  A month is a long time to wait for the next selection pass  Only 25% of ESD is available for event navigating from TAG/AOD during analysis  The 25% of ESD on disk will rarely have been consistently selected (once a month) and will be continuously rotating, altering the accessible subset of data

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 10 Changes in Computing Model and Requirements (2)  Underlying problem:  Selection pass and analysis event navigation access to ESD is sparse  Estimated to be ~1 out of 100 events per analysis  ESD is on tape rather than on disk  Tape is a sequential medium  Must access 100 times more data then needed  Tape is expensive per unit of I/O bandwidth  As much as 10 times that of disk  Thus penalty in access cost relative to disk may be a factor of ~1000  Solution:  Get all ESD on disk  Methods for accomplishing this are:  Buy more disk at Tier 1 – most straight forward  Unify/coordinate use of existing disk across multiple Tier 1’s – more economical  Some combination of above – compromise as necessitated by available funding

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 11 “2007” Capacities for U.S. Tier 1 Options  “3 Tier 1” Model (Complete ESD found on disk of U.S. plus 2 other Tier 1’s)  Highly dependent on the performance of other Tier 1’s and the Grid middleware and network (transatlantic) used to connect to them  “Standalone” Model (Complete ESD on disk of US Tier 1)  While avoiding above dependencies, is more expensive

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 12 Changes in Technology  No dramatic new technologies  Previously assumed technologies are tracking Moore’s Law well  Recent price performance points from RHIC Computing Facility  CPU: IBM procurement - $33/SPECint95  310 Dual 1 GHz Pentium III 97.2 SPECint95/Node  Delivered Aug 2001  $1M fully racked including cluster management hardware & software  Disk: OSSI/LSI procurement - $27k/TByte  33 Usable TB of high availability Fibre Channel RAID 1400 MBytes/sec  Delivered Sept 2001  $887k including SAN switch  Strategy is to project, somewhat conservatively, from these points for facilities design and costing  Actually used 20 month rather than the observed <18 month price/performance halving time for disk and cpu

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 13 Changes in Budgetary Assumptions (2)  Assumed Funding Profiles (At Year $K)  For revise LHC startup schedule, new profile is better  For ATLAS DC 2 which stayed fixed in ’03, new profile is worse  Hardware capacity goals of DC 2 will not be met  Personnel intensive facility development may be as much as 1 year behind  Hope is that another DC will be added allowing validation of a more nearly fully developed Tier 1 and US ATLAS facilities Grid

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 14 Profiles for Standalone Disk Option  Much higher functionality (than other options) and, given new stretched out LHC schedule, within budget guidance  Fractions in revised profiles in table below are of a final system which has nearly 2.5 times the capacity of that discussed last year

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 15 Associated Labor Profile

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 16 Summary Tier 1 Cost Profile (At Year $K)  Current plan violated guidance by $370k in FY ’04, but this is a year of some flexibility in guidance  Strict adherence to FY ’04 guidance would …  reduce facility capacity from 3% to 1.5% or staff by 2 FTE’s

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 17 Tier 1 Capacity Profile

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 18 Tier 1 Cost Profiles

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 19 Standalone Disk Model Benefits  All ESD, AOD, and TAG data on local disk  Enables analysis specific 24 hour selection passes (versus one month aggregated passes) – faster, better tuned, more consistent selection  Allows navigation for individual events (to all processed, but not Raw, data) without recourse to tape and associated delay – faster more detailed analysis of larger consistently selected data sets  Avoids contention between analyses over ESD disk space and the need for complex algorithms to optimize management of that space – better result with less effort  While prepared to serve appropriate levels of data access to other Tier 1’s, US will not in general be unduly sensitive to the performance of other Tier 1’s or intervening network (transatlantic) and middleware – improved system reliability, availability, robustness and performance

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 20 Tier 2 Issues  The high availability of the complete ESD set on disk at the Tier 1 and the associated increased frequency of ESD selection passes will, for connected Tier 2’s (and Tier 3’s ), lead to …  More analysis activity – (Increasing CPU & Disk utilization)  More frequent analysis passes on  More and larger usable TAG, AOD and ESD subsets  More network traffic into the site from the Tier 1 – (Increasing WAN utilization)  Selection results  Event navigation into the full disk resident ESD  As in the case of the Tier 1, an additional year of funding before turn-on and the increased effectiveness of “year later” funding contribute to satisfying these increased needs within or near the integrated out year (’05-’07) budget guidance  The delay of some ’06 funding to ’07 is required for a better match of profiles

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 21 Tier 2 Distribution Of Hardware Cost

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 22 Tier 1 Distribution Of Hardware Cost

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 23 FY 2007 Capacity Comparison of Models

27-30 November, 2001 B. Gibbard Review of US LHC Software and Computing Projects 24 Conclusions  Standalone disk model  A dramatic improvement over previous tape based mode – Functionality & Performance  A significant improvement over multi-Tier 1 disk model – Performance, Reliability & Robustness  Respects funding guidance in model sensitive out-years  If costs are higher or funding lower than expected, a graceful fallback is to access some of the data on disks at other Tier 1’s  Adiabaticly move toward multi-Tier 1 model