Abstract

A novel probabilistic framework for the study of performance issues in multiprocessor scheduling of tasks with uncertain characteristics is proposed. It enables the determination of various performance measures, such as the overall completion rate from a probabilistic picture of the tasks currently under execution, formed from a probabilistic description of newly arriving tasks and assignment of processors. Once optimised, these measures may be used to guide the decisions of schedulers at run time.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call