This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Scheduling despite inexact job-size information Adam Wierman California Institute of Technology 1200 E. California Blvd. Pasadena, CA 91125 [email protected] Misja Nuyens Statkraft Lilleakerveien 6 Lilleaker, 0216 Oslo [email protected] ABSTRACT Motivated by the optimality of Shortest Remaining Process- ing Time ( SRPT ) for mean response time, in recent years many computer systems have used the heuristic of “favoring small jobs” in order to dramatically reduce user response times. However, rarely do computer systems have knowl- edge of exact remaining sizes. In this paper, we introduce the class of- SMART policies, which formalizes the heuris- tic of “favoring small jobs” in a way that includes a wide range of policies that schedule using inexact job-size infor- mation. Examples of- SMART policies include (i) policies that use exact size information, e.g., SRPT and PSJF , (ii) policies that use job-size estimates, and (iii) policies that use a finite number of size-based priority levels. For many - SMART policies, e.g., SRPT with inexact job- size information, there are no analytic results available in the literature. In this work, we prove four main results: we derive upper and lower bounds on the mean response time, the mean slowdown, the response-time tail, and the conditional response time of - SMART policies. In each case, the results explicitly characterize the tradeoff between the accuracy of the job-size information used to prioritize and the performance of the resulting policy. Thus, the results provide designers an understanding of how accurate job-size information must be in order to achieve desired performance guarantees. 1. INTRODUCTION Job-size information is a useful tool for improving the per- formance of schedulers. It has long been known that Short- est Remaining Processing Time first ( SRPT ) scheduling min- imizes mean response time (sojourn time)  and is near optimal for weighted response time measures such as mean slowdown (stretch) . Similarly, variants of SRPT such as Preemptive Shortest Job First ( PSJF ) are known to provide near optimal mean response time and mean slowdown. However, the adoption of designs based on SRPT and PSJF has been slow due to fears about the fairness of these policies. Specifically, there are worries that large job sizes Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee....
View Full Document
- Fall '09
- Greatest element, response time, SRPT