An algorithm for scheduling prioritized tasks in a hard real-time environment

Abstract

This paper addresses the problem of dynamically scheduling a set of prioritized tasks in a Hard Real-Time environment. Since optimal solutions have exponential complexity in the general case, our solution relies on a heuristic search over a set of prioritized tasks. The most important aspect of the proposed algorithm is that it incorporates the priority ordering of the task set while exploring the search tree, Our main objective is to evaluate, through event-driven simulation, a scheme that integrates priority ordering in the search by considering the priority of a task at every level of the search tree. Results show that our approach provides the short and long term adaptability required in large dynamic systems, and overcomes some of the limitations of previously proposed solutions.