IRMM_ROOT_PROOF5天.pdfVIP

Dr Lorenzo Moneta CERN PH-SFT CH-1211 Geneva 23 sftweb.cern.ch root.cern.ch Introduction to PROOF and PROOF Lite 1 ROOT Training at IRMM 1st March 2013 sftweb.cern.ch root.cern.ch ROOT Training at IRMM: Day 5 - Introduction to PROOF 2 Does the load on your 8-core machine look like this during your analysis session? sftweb.cern.ch root.cern.ch ROOT Training at IRMM: Day 5 - Introduction to PROOF 3 What would be needed to make it look like this? sftweb.cern.ch root.cern.ch ROOT Training at IRMM: Day 5 - Introduction to PROOF 4 Just one extra command TProof::Open(“”) sftweb.cern.ch root.cern.ch ROOT Training at IRMM: Day 5 - Introduction to PROOF What is PROOF? ? A system for running ROOT queries in parallel on a large number of distributed computers or many-core machines. ? PROOF is designed to be a transparent, scalable and adaptable extension of the local interactive ROOT analysis session. ? Extends the interactive model to long running “interactive batch” queries. ? Uses xrootd for data access and communication infrastructure. ? For optimal CPU load it needs fast data access (SSD, disk, network) as queries are often I/O bound. ? Can also be used for pure CPU bound tasks like toy Monte Carlo’s for systematic studies or complex fits. 5 sftweb.cern.ch root.cern.ch ROOT Training at IRMM: Day 5 - Introduction to PROOF The PROOF Approach 6 File catalog Master Scheduler Storage CPU’s Query PROOF query: data file list, mySelector.C Feedback, merged final output PROOF cluster ? Cluster perceived as extension of local PC ? Same macro and syntax as in local session ? More dynamic use of resources ? Real-time feedback ? Automatic splitting and merging sftweb.cern.ch root.cern.ch ROOT Training at IRMM: Day 5 - Introduction to PROOF Multi-Tier Architecture 7 Adapts to wide area virtual clusters Geographically separated domains, heterogeneous machines Network performance Less important VERY important Optimize for data locality or high bandwidth data server ac