Distributed Analysis on Panda

Introduction

Setup

First, please make sure you have a grid certificate. See Starting on the Grid. You must have usercert.pem and userkey.pem under the ~/.globus directory.

$ ls ~/.globus/*
usercert.pem userkey.pem

Then, you need to install the panda-client package which contains client tools to submit analsis jobs to PanDA. See the panda-client page for installation and setup procedures. e.g., at CERN, after setting up athena, simply do :

$ source /afs/cern.ch/atlas/offline/external/GRID/DA/panda-client/latest/etc/panda/panda_setup.[c]sh

How to use individual tool

Once you install the panda-client package, the following tools are available.

• pathena
  • How to submit Athena jobs to PanDA
• prun
  • How to submit ROOT/General jobs to PanDA
• psequencer
  • How to perform sequential jobs/operations in PanDA based analysis
• pbook
  • Bookkeeping for PanDA analysis jobs
• puserinfo
  • Access Control on PanDA analysis jobs

Please refer each HowTo page according to your purpose.

Troubleshooting

Please submit all your help requests to the Distributed Analysis Help HyperNews (hn-atlas-dist-analysis-help@cern.ch , sign up to the hn-atlas-dist-analysis-help e-group) which is maintained by AtlasDAST. This forum takes care of site problems (e.g. SE problems) as well. PanDA Jira is for software bug reporting. For general info see BigPanDA monitor.

Contact Email Address: hn-atlas-dist-analysis-help@cern.ch

Responsible:

MassimoLamanna - 5 Sep 2008