An Approximate Analysis of the Shortest Queue Policy on Soft Real-Time Scheduling Weiping Zhu School of C o m p u t e r Science T h e U n i v e r s i t y of New S o u t h W a l e s , A D F A Australia, ACT2600 E-mail weiping@cs.adfa.edu.au Abstract The join the shortest queue (JSQ) policy is studied in the context of soft real-time scheduling. An approximate analysis of the JSQ is developed and presented in this paper. The result obtained from the approximate analysis is compared against the simulation one, that shows the approximate analysis is highly accurate. Thus, the approximate analysis can be applied to the development of soft real-time systems. Introduction With the advent of telecommunication and microprocessor technology, clusters consisting of a number of computers connected by a fast local area network, such as fast Ethernet or Myrinet, have been increasingly used to deliver high performance services [1]. Studies also show the possibility and flexibility of using clusters to provide soft real-time service [2, 3]. To minimize the number of tasks that miss their deadlines, arrival tasks need to be evenly distributed to
/lp/association-for-computing-machinery/an-approximate-analysis-of-the-shortest-queue-policy-on-soft-real-time-A929llGRDm