Infrastructure Monitoring for Distributed Tier1: The ReCaS Project Use-Case

Abstract

The management of distributed e-infrastructures introduces additional requirements in the monitoring system. The main issue is to enable users to achieve the awareness of the global status, and to present the information in function of the role: simple user, system admin, manager. In this work we present the solution created for the ReCaS infrastructure that federates four datacenters in south of Italy. Thanks to the introduction of a data exchange schema, we integrated four heterogeneous and independent subsystems into a single monitoring dashboard. The data collection system allowed us to obtain aggregate metrics by giving a full instantaneous overview of the global infrastructure. Finally, thanks to the user profiling different access views are available, associated to the different class of consumers.

Tools for Distributed Systems Monitoring

“…Monitoring solution has become an inherent part of distributed systems [2]. The main advantage it provides is early failure identification and notification of support teams through email or text message.…”