xAOD monitoring

Introduction

Finally we are in position to get information from most of the analysis code run. These are some of the questions we want to answer:
  • Which data formats are used for analysis?
  • Which branches are used?
  • Difference between on-grid/off-grid?
  • Which versions of ROOT are used?
  • Measure IO (read) performance

Data collection

  • accessedFiles - full path for all the files in the job
  • accessedBranches - for each branch that was accessed a number of accesses/events
  • accessedContainers - for each container that was accessed a number of accesses/events
  • ReadCalls
  • ReadSize
  • CacheSize
  • ROOT_RELEASE
  • pandaID, taskID - if grid job

Data flow

Codes

Please use the official git repository.

Docs

https://docs.google.com/presentation/d/1WhM1aen3N3VQRUB78HlADv6OavLbEpkBsF0xMtHYQyQ/edit?usp=sharing


Major updates:
-- IlijaVukotic - 2015-11-04

Responsible: IlijaVukotic
Last reviewed by: Never reviewed


This topic: AtlasComputing > WebHome > AtlasComputing > AtlasDistributedComputing > ATLASAnalytics > XAODmonitoring
Topic revision: r2 - 2016-05-16 - IlijaVukotic
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2020 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback