A globally distributed grid monitoring system to facilitate high-performance computing at DO/SAM-grid (design, development, implementation and deployment of a prototype).

摘要

A grid environment involves large scale sharing of resources that are distributed from a geographical or an administrative perspective. There is a need for systems that enable continuous discovery and monitoring of the components of a grid. In this work, we discuss the development and deployment of a monitoring system that has been designed as a prototype for the DØ/SAM-Grid. We have developed a system that uses a layered architecture for information generation and processing, utilizes the various grid middleware tools, and implements Integration and Enquiry Protocols using existing Discovery Protocols to provide a user with a coherent view of all current activity in this grid—in the form of a web portal interface. The prototype system has been deployed for monitoring of 11 sites geographically distributed in 5 countries across 3 continents. This work focuses on the DØ/SAM-Grid, and is based on the SAM system developed at Fermilab.