Priority Assignment on Partitioned Multiprocessor Systems With Shared Resources

Driven by industry demand, there is an increasing need to develop real-time multiprocessor systems which contain shared resources. The Multiprocessor Stack Resource Policy (MSRP) and resource sharing Protocol (MrsP) are two major protocols that manage access Both of them can be applied Fixed-Priority Preemptive Scheduling (FPPS), enforced most commercial regulations, requires task priorities assigned before deployment. Along with MSRP MrsP, exist forms schedulability tests bound the worst-case blocking time due accesses: traditional ones being more widely adopted recently developed holistic deliver tighter analysis. On uniprocessor systems, several well-established optimal priority assignment algorithms. Unfortunately, on resources, issue has not been adequately understood. In this article, we investigate three mainstream algorithms-Deadline Monotonic Priority Ordering (DMPO), Audsley's Optimal Assignment (OPA), Robust (RPA), in context partitioned Our contributions multifold: First, prove DMPO tests. Second, counter examples given as evidence Third, then analyze pessimism arising from adoption OPA RPA Lastly, propose a Slack-based (SPO) algorithm minimises such pessimism, polynomial complexity. Comprehensive experiments show SPO outperforms (i.e., results larger number schedulable systems) DMPO, OPA, general tests, up 15 percent. With theoretical contributions, paper useful guide