Intelligent Rate-monotonic Scheduling Algorithm for Real-time Systems

One of the most widely used static priority scheduling algorithms for hard real-time systems serving periodic tasks is the rate-monotonic algorithm. It bene®ts from a rich theoretical background and simplicity of implementation. Since it was ®rst introduced, many researchers have studied its behavior and many practical safety veri®cation algorithms have been developed. In this paper, a modi®ed version of the traditional rate-monotonic algorithm called "preemption intelligent rate-monotonic" is introduced and examined. We have proved that a system of two periodic tasks is safe if and only if the processor load factor is not greater than one. We have also proved that any system that is safe with the rate-monotonic algorithm is also safe with the intelligent rate-monotonic algorithm. Both algorithms were simulated and it was shown that, on the average, for an unsafe system the number of request overruns under the rate-monotonic algorithm is more than that of the intelligent rate-monotonic algorithm. Another result was that the average number of context switchings under the intelligent rate-monotonic algorithm is also less than that of the rate-monotonic algorithm. Reduction in the number of context switchings in turn decreases overhead time.

[1]  Shigemi Aoyagi,et al.  A Feasibility Decision Algorithm for Rate Monotonic and Deadline Monotonic Scheduling , 1998, Real-Time Systems.

[2]  Shirish S. Sathaye,et al.  Generalized rate-monotonic scheduling theory: a framework for developing real-time systems , 1994, Proc. IEEE.

[3]  Miroslaw Malek,et al.  Minimum Achievable Utilization for Fault-Tolerant Processing of Periodic Tasks , 1998, IEEE Trans. Computers.

[4]  Krithi Ramamritham,et al.  Hard Real-Time Systems , 1988 .

[5]  John P. Lehoczky,et al.  Fixed priority scheduling of periodic task sets with arbitrary deadlines , 1990, [1990] Proceedings 11th Real-Time Systems Symposium.

[6]  K. H. Kim,et al.  Prevention of task overruns in real-time non-preemptive multiprogramming systems , 1980, International Symposium on Computer Modeling, Measurement and Evaluation.

[7]  Alan Burns,et al.  An extendible approach for analyzing fixed priority hard real-time tasks , 1994, Real-Time Systems.

[8]  Rami G. Melhem,et al.  Fault-Tolerant Rate-Monotonic Scheduling , 1998, Real-Time Systems.

[9]  James W. Layland,et al.  Scheduling Algorithms for Multiprogramming in a Hard-Real-Time Environment , 1989, JACM.

[10]  Lui Sha,et al.  Real-time scheduling theory and Ada , 1990, Computer.

[11]  John P. Lehoczky,et al.  The rate monotonic scheduling algorithm: exact characterization and average case behavior , 1989, [1989] Proceedings. Real-Time Systems Symposium.

[12]  Joseph Y.-T. Leung,et al.  On the complexity of fixed-priority scheduling of periodic, real-time tasks , 1982, Perform. Evaluation.