Enabling data analysis à la PROOF on the Italian ATLAS Tier-2s using PoD

Abstract

We describe our experience using PROOF for data analysis on the Italian ATLAS-Tier2 in Frascati, Napoli and Roma1. To enable PROOF on the cluster we used PoD, Proof-on-Demand. PoD is a set of tools designed to interact with any Local Resource Management System (LRMS) to start the PROOF daemons. In this way any user can quickly setup its own PROOF cluster on the resources, with the LRMS taking care of scheduling, priorities and accounting. Usage of PoD has steadily increased in the last years, and the product has now reached a production level quality. PoD features an abstract interface to LRMSs and provides plugins for several LRMSs. In our tests we used both the gLite and PBS plug-ins, the latter being the native LRMS handling the resources under test. Data were accessed via XRootD with file discovery provided by the standard ATLAS tools. The Storage Element was Disk Pool Manager (DPM) which traditionally uses RFIO rfio data access protocol; we added XRootD on top of this system so PoD could access the data. We describe the configuration and setup details and the results of some benchmark tests we run on the facility.