The provider, when accepting jobs, has to enqueue them using an agreed scheduling policy. However, the provider also has to keep in mind the optimization of the usage of resources, and the QoS guaranteed in the SLA. This best match has to take into account potential arrival of new demands, already running jobs, the respective priorities, which should deliver the most suitable resources fulfilling the requirements. A prototype algorithm may be proposed.
© 2007 BEinGRID