Monitoring, Accounting and Automated Decision Support for the ALICE Experiment Based on the MonALISA Framework

The successful administration of a global Data Grid system requires collecting and storing relevant monitoring information, using it to show the status and the trends of the entire system. The collected information is further used in developing the higher-level services and components of the distributed system to provide a degree of automated operational decisions, in order to maintain and optimize the work-flow through the entire system. In this paper we present the architecture of the monitoring system, developed within the MonALISA framework, for the ALICE Grid. The system uses flexible mechanisms for collecting, aggregating, storing and presenting monitoring information both in near real-time or history charts, in global or specific views, being able to generate alerts or take automated decisions based on it