On Optimal Priority Assignment for Response Time Analysis of Global Fixed Priority Pre-emptive Scheduling in Multiprocessor Hard Real-Time Systems

This paper investigates the problem of optimal priority assignment in multiprocessor real-time systems using global fixed task-priority pre-emptive scheduling. Previous work in this area showed that arguably the most effective pseudo-polynomial schedulability tests for global fixed priority pre-emptive scheduling, based on response time analysis, are not compatible with Audsley’s Optimal Priority Assignment (OPA) algorithm. In this paper, we derive upper and lower bounds on these response time tests that are compatible with the OPA algorithm. We show how these bounds can be used to limit the number of priority ordering combinations that need to be examined, and thus derive an optimal priority assignment algorithm with backtracking that is compatible with response time analysis. We show that response time analysis combined with the OPA-backtracking algorithm dominates previous approaches using OPA-compatible polynomial-time schedulability tests.