1 / 5

GURB

GURB. A tool for grid distributed analysis at IFAE. Carlos Osuna. Atlas Ifae meeting - 15 Sept 2006. GURB is a set of tools to manage massive jobs to send to the grid for IFAE users (user should be familiar with grid environment). PIC batch system will not be available much longer.

zagiri
Télécharger la présentation

GURB

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. GURB A tool for grid distributed analysis at IFAE Carlos Osuna Atlas Ifae meeting - 15 Sept 2006

  2. GURB is a set of tools to manage massive jobs to send to the grid for IFAE users (user should be familiar with grid environment). PIC batch system will not be available much longer for big jobs, grid is the only possible computing solution for ATLAS ifae Need to have a good tool to manage high number of jobs in the grid official distributed analysis tools for atlas still in development (not flexible enough) GANGA missing desired funcionalities. PANDA needs to contact central "executor"--> slow!! manage big jobs not easy, at least more complicated than batch: where are user libraries (in home area in batch)? where are jobOptions? additional scripts? how to get input files? from where? how to retrieve output files? if big enough, what to do with them? GURB was written in python, I hope any user can debug it, add new features to the tool Documentation available at: http://twiki.cern.ch/twiki/bin/view/Main/IfaeAtlasTier2 small tutorial could be given in case people request...

  3. get user libs grid send n jobs PIC lfn.rome.0485......_0009.pool.root lfn.rome.0485......_0009.pool.root lfn.rome.0485......_0009.pool.root .............. lfn.rome.0485......_0009.pool.root lfn.rome.0485......_0009.pool.root Split job (gridify) Register user libs

  4. running jobs register output (if too big to be retrieved) grid retrieve output PIC

  5. grid PIC lfn.rome.0485......_0009.pool.root lfn.rome.0485......_0009.pool.root lfn.rome.0485......_0009.pool.root .............. lfn.rome.0485......_0009.pool.root lfn.rome.0485......_0009.pool.root repeat process with jobs that failed previously

More Related