1 / 31

The HENP Grand Challenge Project and initial use in the RHIC Mock Data Challenge 1

The HENP Grand Challenge Project and initial use in the RHIC Mock Data Challenge 1. D. Olson DM Workshop SLAC, 20-22 Oct 1998. Outline. Overview The problem being addressed Experiences from Mock Data Challenge. The HENP GCA. 3 year project: FY97, FY98, FY99

adrina
Download Presentation

The HENP Grand Challenge Project and initial use in the RHIC Mock Data Challenge 1

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. The HENP Grand Challenge Projectand initial use in the RHIC Mock Data Challenge 1 D. Olson DM Workshop SLAC, 20-22 Oct 1998

  2. Outline • Overview • The problem being addressed • Experiences from Mock Data Challenge HENP-GC, D. Olson, SLAC DM Workshop

  3. The HENP GCA • 3 year project: FY97, FY98, FY99 • Funding from DOE/MICS, collaboration with DOE/NP, HEP • Focus on RHIC data access HENP-GC, D. Olson, SLAC DM Workshop

  4. Who - the workers • Henrik Nordberg, NERSC/LBNL • Luis Bernardo, NERSC/LBNL • Alex Sim, NERSC/LBNL • Dave Malon, ATLAS/ANL • Dave Stampf, RCF/BNL • Jeff Porter, STAR/LBNL • Dave Zimmerman, STAR/LBNL • Jie Yang, STAR/LBNL-UCLA-Beijing • Mark Pollack, PHENIX/BNL HENP-GC, D. Olson, SLAC DM Workshop

  5. Who - the others • Doug Olson - STAR/LBNLArie Shoshani, Doron Rotem - NERSC/LBNL (Data Mgmt Grp)Craig Tull - NERSC/LBNL (HENP) • Bruce Gibbard, Shigeki Misawa RCF/BNLTorre Wenaus STAR/BNL • ED May - ATLAS/ANL HENP-GC, D. Olson, SLAC DM Workshop

  6. Relativistic Heavy Ion Collider • Brookhaven National Laboratory on Long Island • An accelerator for high-energy nuclear physics • Begin operating in June 1999. (10+ year life) HENP-GC, D. Olson, SLAC DM Workshop

  7. Using ROOT (root.cern.ch) “small” Using Objectivity/DB (www.objectvity.com) “large” 2 “Large”, 2 “Small” Experiments(www.rhic.bnl.gov) HENP-GC, D. Olson, SLAC DM Workshop

  8. Characteristics BRAHMS PHENIX PHOBOS STAR # Scientists (approx.) 50 400 70 350 # Institutions 13 45 12 36 M events/year 3,600 965 2880 17 Size/raw event (KB) 10 300 18 12000 Total Data/Year (TB) 62 496 204 264 Req'd CPU Capacity 960 17,518 6,196 8,818 (SPECint95) HENP-GC, D. Olson, SLAC DM Workshop

  9. Different event components stored in different files. Event (data) structure for STAR Tags (index) Event components, (bulky data) http://www.rhic.bnl.gov/STAR/html/comp_l/dataproc/EventStructure.pdf HENP-GC, D. Olson, SLAC DM Workshop

  10. Every user is also a software developer. Data Characteristics (STAR example) http://www.rhic.bnl.gov/STAR/html/comp_l/ofl/reqmts9708/report/CompReqReport.ps HENP-GC, D. Olson, SLAC DM Workshop

  11. Certainly Probably Possibly Doubtful Likelihood of implementation w/ Objectivity/DB HENP-GC, D. Olson, SLAC DM Workshop

  12. Processor Cache memory mgr Registers CPU cache “I/O” software (Objectivity, Zebra, …) RAM Disk HPSS, pftp (optimization with HENP GC) Tape Robot Shelf “Hope for” HPSS Transport through the storage hierarchy HENP-GC, D. Olson, SLAC DM Workshop

  13. The Goal • Optimize access to tape-resident files • Based upon selections of objects of interest to the application (components of physics events) • Utilizing disk-resident index HENP-GC, D. Olson, SLAC DM Workshop

  14. RHIC Analysis Architecture MDC2 MDC2 HENP-GC, D. Olson, SLAC DM Workshop

  15. HENP-GC software features • Index event component objects • Query attributes of events (tags) • Order optimize iteration over events • Coordinate file caching across multiple simultaneous queries • Policies to control resource usage • Parallel query execution (analysis) HENP-GC, D. Olson, SLAC DM Workshop

  16. Opportunities for optimization • Prevent / eliminate unwanted queries=> query estimation (fast index) • Read all events (qualified for a query) from a file at the same time, without reading all event in the file=> exact index over all properties • Share files brought into cache by multiple queries=> look ahead for files needed and cache management • Match data storage to access patterns=> clustering on tape HENP-GC, D. Olson, SLAC DM Workshop

  17. Data access s/w (simple view) GC system components Sample User Code • key developers • Henrik Nordberg (NERSC)query estimator • Alex Sim (NERSC)query monitor • Luis Bernardo (NERSC)cache manager • Jeff Porter (LBL-STAR)query object • Dave Malon (ANL)order-optimized iterator &gcaResources API • Dave Zimmerman, (LBL-STAR)Mark Pollack (BNL-PHENIX)tagDB • Jie Yang (UCLA,LBL,Beijing)testing Query Interface Storage Manager Event Data (Objectivity) Expt. code & data HENP-GC, D. Olson, SLAC DM Workshop

  18. Query Estimator Event Iterators Process Flow 1 execute 7 retrieve whichFileToCache 3 Query Monitor Policy Module 4 2 request FileID ToCache 8 release 5 stage 6 staged 9 purge 10 purged Cache Manager HENP-GC, D. Olson, SLAC DM Workshop

  19. MDC1 setup STAR Objectivity database files in PHENIX COS, 2 tapes STAR Objectivity database files in STAR COS, 2 tapes, 32 GB, 240 files Storage manager and analysis codes on rmds03 Objy db files on local disk pftp pftp HENP-GC, D. Olson, SLAC DM Workshop

  20. Legend File staged to local disk File staged to HPSS disk Event ID’s retrieved by iterator Start of pftp request File released by queries Start of query File purged from disk File ID End of query Symbol color identifies query Time HENP-GC, D. Olson, SLAC DM Workshop

  21. 3 queries q3 q1 q2 q1,2,3 3 queries with some shared files, time delay between each query, then the same 3 queries are repeated simultaneously. The cache was large enough to hold all files so the second time all queries run at processing speed rather than I/O speed. HENP-GC, D. Olson, SLAC DM Workshop

  22. Shared access policy With caching policy - shared files, ordering by # events No caching policy HENP-GC, D. Olson, SLAC DM Workshop

  23. Detail Green means pftp failed HPSS recovered & pftp succeeds again HENP-GC, D. Olson, SLAC DM Workshop

  24. Implementation Opportunities for optimization • Prevent / eliminate unwanted queries=> query estimation (fast index) • Query Estimator • Read all events (qualified for a query) from a file at the same time, without reading all event in the file=> exact index over all properties • Order Optimized Iterator • Share files brought into cache by multiple queries=> look ahead for files needed and cache management • Query Monitor • Match data storage to access patterns=> clustering on tape • Clustering Analyzer and Dynamic Reorganizer HENP-GC, D. Olson, SLAC DM Workshop

  25. Things not discussed • Indices • Cluster analysis • Reorganization • Parallel query execution • Cray T3E production of simulated data HENP-GC, D. Olson, SLAC DM Workshop

  26. References • http://www-rnc.lbl.gov/GC/ • http://gizmo.lbl.gov/sm/ • http://www.rhic.bnl.gov/RCF/ • http://www.rhic.bnl.gov/STAR/ HENP-GC, D. Olson, SLAC DM Workshop

  27. The End HENP-GC, D. Olson, SLAC DM Workshop

  28. Where • Massive simulations data generation • NERSC Cray T3E (www.nersc.gov) • Pittsburg SC Cray T3E (www.psc.edu) • Software development & testing • NERSC HPSS, PDSF(recently upgraded from SSC vintage) • Installation & operations • RHIC Computing Facility • STAR regional facility at NERSC/PDSF HENP-GC, D. Olson, SLAC DM Workshop

  29. When • Started March ‘97 • Architecture November ‘97 • RHIC Objectivity decision November ‘97 • Prototype components May ‘98 • RHIC MDC1 September ‘98 • RHIC MDC2 early ‘99 • RHIC operations start November ‘99 HENP-GC, D. Olson, SLAC DM Workshop

  30. Features (MDC1) • Extract tag parameters for index • base attributes & computed values • Query estimation • # events, # files (disk, tape), # seconds • Query execution • order optimization (sort OID’s by file) • return OID’s as files are staged • Disk cache management • pre-fetch files • coordinate multiple queries HENP-GC, D. Olson, SLAC DM Workshop

  31. FY99 • Multi-component event implementation (MDC2) • Performance measurements • Monitoring • Tuning with policy module parameters • GUI’s for • user query builder • administration HENP-GC, D. Olson, SLAC DM Workshop

More Related