Abstract
The development of supportive middleware to manage resources and distributed workload across multiple administrative boundaries is of central importance to Grid computing. Active middleware services that perform look-up, match-making, scheduling and staging are being developed that allow users to identify and utilise appropriate resources that provide sustainable system- and user-level qualities of service.
This paper documents two performance-responsive middleware services that address the implications of executing a particular workload on a given set of resources. These services are based on an estab- lished performance prediction system that is employed at both the local (intra-domain) and global (multi- domain) levels to provide dynamic workload steering. These additional facilities bring about significant performance improvements, the details of which are presented with regard to the user-perceived quality of service and to resource utilisation.