Near Real Time Reconstruction of PHENIX Run7 Minimum Bias Data From RHIC Project Goals Reconstruct 10% of PHENIX min bias data from the RHIC Run7 (Spring.

Slides:



Advertisements
Similar presentations
Buffers & Spoolers J L Martin Think about it… All I/O is relatively slow. For most of us, input by typing is painfully slow. From the CPUs point.
Advertisements

23/04/2008VLVnT08, Toulon, FR, April 2008, M. Stavrianakou, NESTOR-NOA 1 First thoughts for KM3Net on-shore data storage and distribution Facilities VLV.
Data Management for Physics Analysis in PHENIX (BNL, RHIC) Evaluation of Grid architecture components in PHENIX context Barbara Jacak, Roy Lacey, Saskia.
I/O Hardware n Incredible variety of I/O devices n Common concepts: – Port – connection point to the computer – Bus (daisy chain or shared direct access)
Large scale data flow in local and GRID environment V.Kolosov, I.Korolko, S.Makarychev ITEP Moscow.
Data oriented job submission scheme for the PHENIX user analysis in CCJ Tomoaki Nakamura, Hideto En’yo, Takashi Ichihara, Yasushi Watanabe and Satoshi.
Alexandre A. P. Suaide VI DOSAR workshop, São Paulo, 2005 STAR grid activities and São Paulo experience.
03/27/2003CHEP20031 Remote Operation of a Monte Carlo Production Farm Using Globus Dirk Hufnagel, Teela Pulliam, Thomas Allmendinger, Klaus Honscheid (Ohio.
Alain Romeyer - 15/06/20041 CMS farm Mons Final goal : included in the GRID CMS framework To be involved in the CMS data processing scheme.
D0 Farms 1 D0 Run II Farms M. Diesburg, B.Alcorn, J.Bakken, T.Dawson, D.Fagan, J.Fromm, K.Genser, L.Giacchetti, D.Holmgren, T.Jones, T.Levshina, L.Lueking,
An Overview of PHENIX Computing Ju Hwan Kang (Yonsei Univ.) and Jysoo Lee (KISTI) International HEP DataGrid Workshop November 8 ~ 9, 2002 Kyungpook National.
CHEP Sep Andrey PHENIX Job Submission/Monitoring in transition to the Grid Infrastructure Andrey Y. Shevel, Barbara Jacak,
14 Aug 08DOE Review John Huth ATLAS Computing at Harvard John Huth.
PHENIX and the data grid >400 collaborators Active on 3 continents + Brazil 100’s of TB of data per year Complex data with multiple disparate physics goals.
BNL Wide Area Data Transfer for RHIC & ATLAS: Experience and Plans Bruce G. Gibbard CHEP 2006 Mumbai, India.
CC-J Monthly Report Shin’ya Sawada (KEK) for CC-J Working Group
February 28, 2003Eric Hjort PDSF Status and Overview Eric Hjort, LBNL STAR Collaboration Meeting February 28, 2003.
PC clusters in KEK A.Manabe KEK(Japan). 22 May '01LSCC WS '012 PC clusters in KEK s Belle (in KEKB) PC clusters s Neutron Shielding Simulation cluster.
PHENIX Simulation System 1 January 12, 2000 Simulation: Status for VRDC Tarun Ghosh, Indrani Ojha, Charles Vanderbilt University.
PHENIX and the data grid >400 collaborators 3 continents + Israel +Brazil 100’s of TB of data per year Complex data with multiple disparate physics goals.
HIGUCHI Takeo Department of Physics, Faulty of Science, University of Tokyo Representing dBASF Development Team BELLE/CHEP20001 Distributed BELLE Analysis.
University user perspectives of the ideal computing environment and SLAC’s role Bill Lockman Outline: View of the ideal computing environment ATLAS Computing.
RCF Status - Introduction PHENIX and STAR Counting Houses are connected to RCF at a Network Bandwidth of 20 Gbits/sec each –Redundant (Bandwidth-wise and.
November 10, 1999PHENIX CC-J Updates in Nov.991 PHENIX CC-J Updates in Nov New Hardware - N.Hayashi / RIKEN November 10, 1999 PHENIX Computing Meeting.
May 10, 2000PHENIX CC-J Updates1 PHENIX CC-J Updates - Preparation For Opening - N.Hayashi / RIKEN May 10, 2000 PHENIX Computing
Analyzing ever growing datasets in PHENIX Chris Pinkenburg for the PHENIX collaboration.
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.
6/21/2007PHENIX - MJL1 cosmic ray muons PHENIX Run7 Update Mike Leitch, LANL PHENIX Run Coordinator RHIC/AGS Users Meeting 21 June, 2007 RXNP MPC BBC reaction.
D0 Farms 1 D0 Run II Farms M. Diesburg, B.Alcorn, J.Bakken, R. Brock,T.Dawson, D.Fagan, J.Fromm, K.Genser, L.Giacchetti, D.Holmgren, T.Jones, T.Levshina,
Markus Frank (CERN) & Albert Puig (UB).  An opportunity (Motivation)  Adopted approach  Implementation specifics  Status  Conclusions 2.
ALICE Computing Model A pictorial guide. ALICE Computing Model External T1 CERN T0 During pp run i (7 months): P2: data taking T0: first reconstruction.
CCJ introduction RIKEN Nishina Center Kohei Shoji.
Status Report on Data Reconstruction May 2002 C.Bloise Results of the study of the reconstructed events in year 2001 Data Reprocessing in Y2002 DST production.
1 P. Murat, Mini-review of the CDF Computing Plan 2006, 2005/10/18 An Update to the CDF Offline Plan and FY2006 Budget ● Outline: – CDF computing model.
September 26, 2003K User's Meeting1 CCJ Usage for Belle Monte Carlo production and analysis –CPU time: 170K hours (Aug.1, 02 ~ Aug.22, 03)
Getting the Most out of Scientific Computing Resources
Getting the Most out of Scientific Computing Resources
Module 12: I/O Systems I/O hardware Application I/O Interface
Charles Maguire et (beaucoup d’) al.
Real Time Fake Analysis at PIC
WP18, High-speed data recording Krzysztof Wrona, European XFEL
CMS-HI Offline Computing
Charles Maguire for VU-RHIC group
The CMS-HI Computing Plan Vanderbilt University
Vanderbilt Tier 2 Project
Bernd Panzer-Steindel, CERN/IT
Offline data taking and processing
ALICE Computing Model in Run3
Grid Canada Testbed using HEP applications
Preparations for the CMS-HI Computing Workshop in Bologna
R. Graciani for LHCb Mumbay, Feb 2006
Vanderbilt University
Nuclear Physics Data Management Needs Bruce G. Gibbard
Proposal to BNL/DOE to Use ACCRE Farm for PHENIX Real Data Reconstruction in Charles Maguire for the VU PHENIX group Carie Kennedy, Paul Sheldon,
Preparations for Reconstruction of Run6 Level2 Filtered PRDFs at Vanderbilt’s ACCRE Farm Charles Maguire et al. March 14, 2006 Local Group Meeting.
Operating System Concepts
13: I/O Systems I/O hardwared Application I/O Interface
CS703 - Advanced Operating Systems
Status of CMS-HI Compute Proposal for USDOE
Status of CMS-HI Compute Proposal for USDOE
Michael P. McCumber Task Force Meeting April 3, 2006
Heavy Ion Physics Program of CMS Proposal for Offline Computing
Heavy Ion Physics Program of CMS Proposal for Offline Computing
CMS-HI Offline Computing
RHIC Computing Facility Processing Systems
Chapter 13: I/O Systems I/O Hardware Application I/O Interface
Preparations for Reconstruction of Run7 Min Bias PRDFs at Vanderbilt’s ACCRE Farm (more substantial update set for next week) Charles Maguire et al. March.
The ATLAS Computing Model
Expanding the PHENIX Reconstruction Universe
Module 12: I/O Systems I/O hardwared Application I/O Interface
Presentation transcript:

Near Real Time Reconstruction of PHENIX Run7 Minimum Bias Data From RHIC Project Goals Reconstruct 10% of PHENIX min bias data from the RHIC Run7 (Spring 2007, Au+Au @ 200 GeV) in near real time at Vanderbilt University computer farm Envisioned as a demonstration project to PHENIX, CMS, VU, and DOE Speed the process of deriving physics information from the detector’s raw data Currently about one year elapses between end of data taking and end of reconstruction Run7 data taken by PHENIX may not be analyzed in time for QM’08 (February 2008) In future Vanderbilt could be reconstructing ~1/3 of PHENIX raw data in real time, with ~1/3 analyzed at CCF and CCJ (PHENIX computer centers France and Japan), and 1/3 at RHIC Computer Facility (RCF) ===> more time and more CPUs for analyses ! Project Requirements Large computer farm with ~50 TBytes of disk space, 200 CPUs Fast and automated file transport to and from the Vanderbilt computer farm Close coordination among PHENIX, Vanderbilt, and the RHIC Computer Facility Dedicated scientific staff at Vanderbilt (1 faculty, 1 PD, 3 students) Most important facet is complete automation and comprehensive monitoring tools June 15, 2007 CMS-HI at CERN

PRDFs Vanderbilt Farm 1600 CPUs, 80 TBytes disk 45TB and 200 CPUs Available for Run7 Reconstruction RCF RHIC computing facility Reconstruction 200 jobs/cycle PRDFs nanoDSTs 18 hours/job 3 FDT 45 MB/s FDT 45 MB/s 1 770 GBytes per cycle PRDFs Raw data files GridFTP to VU 30 MBytes/sec 2 nanoDSTs Reco output to RCF GridFTP 23 MB/sec Dedicated GridFTP Server Firebird 4 4.4TB of buffer disk space June 15, 2007

Actual Experience April - June 2007 Raw Data File Transport (PHENIX DAQ -> Vanderbilt) Automated GridFTP tools in use since 2005 (to CCJ, ORNL, VU, CCF) Works very well and fast enough (30 MBytes/second for current needs) Can (and should) be pushed harder New FDT scripts run for internal transfers can run at 100 MBytes/second FDT is a java-based technology supported out of CERN Throttled back to 45 MBytes/second to minimize impact elsewhere on farm May receive 37 TB of raw data (Au+Au, 200 GeV) containing 325M events by the end of June Reconstruction cycle Larger than expected output size: 70% of input size, PHENIX had previewed 35% Transport time of output limits the compute cycle to being 200 jobs to avoid pileup Cycle is 18 hours (jobs are 12 CPU-hours but new quad CPUs are only 75% efficient [memory bus] !!) Weekends, holidays are slack times --> 200 jobs are start running right away Weekdays are busier, means that there can be a 12 hour wait for the last job to start Reconstructed Output File Transport (Vanderbilt -> RCF) Most fragile component due to disk I/O competition at RCF with other PHENIX users PHENIX is severely short on disk space for this year, but better next fiscal year Developed “fault tolerant” GridFTP scripts which recover if transfers fail mid-stream Large demand to view files arriving at RCF (4 new subsystems installed for Run7) June 15, 2007 CMS-HI at CERN