[Previous [1 [2 [3 [4 [5 [6 [7 [8

Journal of Inforamtion Science and Engineering, Vol. 16, No. 3, pp. 333-347 (May 2000)

Real-Time Gang Schedulings With Workload Models for
Parallel Computers*

Wei-Kuan Shih, Chung-Der Lin, Yar-Wen Chang+
and Jenq-Kuen Lee

Department of Computer Science
National Tsing-Hua University
Hsinchu, Taiwan 300, R.O.C.

Gang scheduling has recently been shown to be an effective task scheduling policy for parallel computers because it combines elements of space sharing and time sharing [10, 20]. In this paper, we propose new policies to enable gang scheduling to adapt to environments with real-time constraints. Our work, to our best knowledge, is the first attempt to address these real-time aspects with gang scheduling. Our system, guided by a metric called the task utilization workload,?can schedule both real-time and non-real-time tasks at the same time. In this paper, we report simulation results obtained using a family of scheduling algorithms based on our proposed metric. Our scheme is designed for practical use with large scale industrial and commercial parallel systems. Preliminary simulation results also show that our proposed policy is effective for real-time scheduling and can schedule non-real-time tasks with fairness and good throughput.

Keywords: real-time scheduling, gang-scheduling, parallel softwares, hierarchical and distributed control, multiprocessor scheduling algorithm

Full Text () Retrieve PDF document (200005_02.pdf)

Received April 12, 1999; revised August 4, 1999; accepted October 19, 1999.
Communicated by Pen-Chung Yew.
*This work was supported in part by National Science Council of Taiwan under grant Nos. NSC88-2215-E-007-034, NSC87-2215-E-007-014, NSC87-2213-E-007-027 and NSC86-2213-E-007-043.