1 / 8

PROOF - Parallel ROOT Facility

PROOF - Parallel ROOT Facility. Maarten Ballintijn, Rene Brun, Fons Rademakers, Gunter Roland http://root.cern.ch. Bring the KB to the PB not the PB to the KB. Slave. Slave. Master. Slave. Slave. Internet. PROOF in a Slide.

nuncio
Download Presentation

PROOF - Parallel ROOT Facility

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. PROOF - Parallel ROOT Facility Maarten Ballintijn, Rene Brun, Fons Rademakers, Gunter Roland http://root.cern.ch Bring the KB to the PB not the PB to the KB

  2. Slave Slave Master Slave Slave Internet PROOF in a Slide • Interactive Physics Analysis of Large Data Sets using Parallel Processing and I/O • Transparency, Adaptability, Scalability • Integrated into ROOT, Known Paradigm • Adapt to Network and System Performance, support Heterogeneous systems • Efficiency and Reliability • Multi Tier architecture, Data Locality • WAN Ready and GRID compatible User

  3. PROOF Features • Selector framework • Transparent way to program analysis • Universally applicable, local, PROOF, batch • Extensible, combine with TTask ? • TDSet data description • Lightweight, easily created from CLI • Versatile

  4. PROOF Features • Packages • Extend and customize environment • Efficient and heterogeneous • Feedback • Simple progress bar • Flexible feedback mechanism

  5. PROOF in the Pipeline • Continuing Grid Integration • Authentication ready, Catalogs, Resource Broker • Dynamic Session Configuration Plugin • Requested Data Set, Resource Availability • Feature additions and Improvements • TTree::Draw(), Friend Trees, Event lists • Tuning and improved error handling

  6. PROOF and Condor • COD, Computing On Demand • Batch and Interactive coexist • Explicit scheduling • Better Interactive Response • Optimize resource usage • Resource broker • Allocate Compute resources • Allocate Data Storage resources

  7. Master Condor PROOF and COD Condor Batch Slave Condor Slave Batch Condor Batch

  8. Software Architecture Philosophy • General principles remain valid • even as bandwidth, latency and capacity grow • We must benefit from what we and others (!) have learned • Models of usage • Silver bullets do not exist • Methodology, Environments, Languages • Prototype is easy, production quality is hard

More Related