Grid Based Monitoring on the Rutgers CDF Analysis Farm

Abstract

Run II at the Fermilab Tevatron Collider started in March 2001, and it will continue probing the high energy frontier in particle physics until the start of the LHC at CERN. The CDF collaboration at Fermilab has already stored 260 TB of data and expects to store 1PB of data in the next two years. The HEXCAF computing farm is being set up at Rutgers University to provide the software environment, computing resources, and access to data for physicists participating in the Collaboration. Some job submission, detector data access and storage of the output results are based on the SAM-GRID tools. To extend monitoring for these jobs running on the farm a bridge was developed between the SAM-GRID monitoring tools and the internal farm monitoring. This presentation will describe the configuration and functionality of the HEXCAF farm with the emphasis on the monitoring tools. Finally we summarize our experience of installing and operating a GRID environment on a remote cluster that is being used for real physics studies in the big running experiment.