An Efficient Heuristic Scheduler for Hard Real-Time Systems
Abstract
The requirement for efficient scheduling algorithms for the development of hard real-time systems resulted in much effort directed toward the development of high performance scheduling algorithms. The algorithms developed up to this point for the Computer Aided Prototyping System (CAPS) do not satisfy the requirements for a efficient static scheduling algorithm. The existing static scheduler neither performs efficiently nor produces correct results for all input cases. This thesis represents the research conducted to develop a fast heuristic static scheduling algorithm based on the principles of simulated annealing. In addition, this thesis describes the development of new data structures that simplify the static scheduler and maximize system resources. Several of the existing scheduling algorithms were re-implemented to make use of the new data structures and provide correct results. Any feasible schedule produced by these scheduling algorithms guarantees that both timing and precedence constraints are met. The primary goal of this thesis was to produce an efficient and effective scheduler to support the CAPS system.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 01, 1991
- Accession Number
- ADA247322
Entities
People
- John G. Levine
Organizations
- Naval Postgraduate School