Scheduling and Priority Mapping for Static Real-Time Middleware

Abstract

This paper presents a middleware real-time scheduling technique for static, distributed, real-time applications. The technique uses global deadline monotonic priority assignment to clients and the Distributed Priority Ceiling protocol to provide concurrency control and priorities for server execution. The paper presents a new algorithm for mapping the potentially large number of unique global priorities required by this scheduling technique to the restricted set of priorities provided by commercial real-time operating systems. This algorithm is called Lowest Overlap First Priority Mapping; we prove that it is optimal among direct priority mapping algorithms. This paper also presents the implementation of these real-time middleware scheduling techniques in a Scheduling Service that meets the interface proposed for such a service in the Real-Time CORBA 1.0 standard. Our prototype Scheduling Service is integrated with the commercial PERTS tool that provides schedulability analysis and automated generation of global and local priorities for clients and servers.