You are not logged in.

Concerning the length of time slots for efficient gang scheduling

Zhou, B., Goscinski, Andrzej and Brent, R. P. 2002, Concerning the length of time slots for efficient gang scheduling, in PDCAT 2002 : Third International Conference on Parallel and Distributed Computing, Applications and Technologies : September 3-6, 2002, Kanazawa, Japan, Australian National University, Canberra, A.C.T., pp. 364-371.

Attached Files
Name Description MIMEType Size Downloads

Title Concerning the length of time slots for efficient gang scheduling
Author(s) Zhou, B.
Goscinski, Andrzej
Brent, R. P.
Conference name International Conference on Parallel and Distributed Computing, Applications and Technologies (3rd : 2002 : Kanazawa, Japan)
Conference location Kanazawa, Japan
Conference dates September 3-6, 2002
Title of proceedings PDCAT 2002 : Third International Conference on Parallel and Distributed Computing, Applications and Technologies : September 3-6, 2002, Kanazawa, Japan
Editor(s) Horiguchi, Susumu
Shen, Hong
Publication date 2002
Series PDCAT 2002 : Third International Conference on Parallel and Distributed Computing, Applications and Technologies : September 3-6, 2002, Kanazawa, Japan
Start page 364
End page 371
Publisher Australian National University
Place of publication Canberra, A.C.T.
Summary Applying gang scheduling can alleviate the blockade problem caused by exclusively used space-sharing strategies for parallel processing. However, the original form of gang scheduling is not practical as there are several fundamental problems associated with it. Recently many researchers have developed new strategies to alleviate some of these problems. Unfortunately, one important problem has not been so far seriously addressed, that is, how to set the length of time slot to obtain a good performance of gang scheduling. With gang scheduling time is divided into time slots of equal length, the number of time slots introduced in the system forms a scheduling round and each new job is first allocated to a particular time slot and then starts to run in the following scheduling round. Ideally, the length of time slot should be set long to avoid frequent context switches and so to reduce the scheduling overhead. The number of time slots in a scheduling round should also be limited to avoid a large number of jobs competing for limited resources (CPU time and memory). Long time slots and the limited number of time slots in each scheduling round may cause jobs to wait for a long time before it can be executed after arrival, which can significantly affect the performance of jobs, especially short jobs which are normally expected to finish quickly. However, the performance of a short job can also suffer if the length of time slot is not long enough to let the short job complete in a single time slot. In this paper we present a strategy to deal with this important issue for efficient gang scheduling.
ISBN 4990033027
9784990033026
Language eng
Field of Research 080307 Operating Systems
HERDC Research category E1 Full written paper - refereed
Copyright notice ©2002 The Authors
Persistent URL http://hdl.handle.net/10536/DRO/DU:30004697

Document type: Conference Paper
Collection: School of Information Technology
Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 0 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 525 Abstract Views, 0 File Downloads  -  Detailed Statistics
Created: Mon, 07 Jul 2008, 09:40:21 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.