Improvement of the processors operating ratio in task scheduling using the deadline method
Task scheduling technique which allocates some tasks to some processors is essential to high performance computing. Scheduling to the processors is crucial for optimizing performance. The objective of scheduling is to minimize the overall completion time or schedule length of the parallel program. On the other hand, a processors operating ratio may fall with the algorithm which pursued only this purpose unfortunately. For improvement of a processors operating ratio, there are the limitation method and the deadline method which we have proposed. In those methods, we limit the number of available processors. In this paper, we propose the method of improving the deadline method, by changing the limitation value of the number of available processors.
Keywordstask scheduling processors operating ratio multi-processor system parallel processing
Unable to display preview. Download preview PDF.
- K. Kashiwagj and S. Kobayashi, “Limitation of used processor for task scheduling,” Proc. 19st IASTED International Multi-Conf. Applied Informatics, pp.122–126, Innsbruck, Austria, Feb. 2001.Google Scholar
- K. Kashiwagj and S. Kobayashi, “Consideration of task’s deadline for cheduling method with used processors limitation,” Proc. Advanced Computer Systems, no. VTI-9, pp.487–496, Memo, Poland, Oct. 2001.Google Scholar
- H. Kasahara and S. Narita, “Practical multiprocessor scheduling algorithms for efficient parallel processing,” IEEE Trans. Computers, no.33(11), pp. 1023–1029, 1984.Google Scholar
- Albert Y. Zomaya, Parallel & distributed computing handbook, McGraw-Hill, New York, 1996.Google Scholar
- Thomas H. Cormen, Charles E. Leiserson, Ronald L. Rivest, Introduction to algorithms, The MIT Press, Massachusetts, 1994.Google Scholar
- H. Hashimoto, K. Kashiwagi, Y. Higami and S. Kobayashi, “How to limit the number of available processors for every task in scheduling”, Proc. SJCIEE2001 (Japanese), Sept. 2001.Google Scholar