Condor: a distributed job scheduler
Beowulf cluster computing with Linux
Distributed computing in practice: the Condor experience: Research Articles
Concurrency and Computation: Practice & Experience - Grid Performance
Hi-index | 0.00 |
Scheduling services are core grid components of paramount importance to support the transparent distribution of tasks to remote shared resources in an efficient way. High availability of these core services is thus of great importance. Given the distributed nature of the system, monitoring the task lifecycle and the aggregate workflow patterns generated by users belonging to various communities is particularly challenging. This paper deals with the problem of grid workload monitoring by reviewing the related requirements, and illustrates the architecture and implementation of a tool, the WMSMonitor, which is designed to meet the needs of various users categories, such as administrators, developers, advanced grid users and performance testers.