Performance Analysis of Parallel Job Scheduling in Distributed Systems
Abstract
This paper studies parallel job scheduling in adistributed system. A simulation model is used to addressperformance issues associated with scheduling. Fivepolicies are used to schedule parallel jobs over a varietyof workloads. Fairness is required among competing jobs.We examine a case where the distribution of the numberof parallel tasks per job and also the distribution of taskservice demand vary with time. Simulated results indicatethat although all scheduling methods have merit, onemethod significantly improves the overall performanceand also guarantees fairness in terms of individual jobexecution.
References
[1]
{1} Bolch, G., S. Greiner, H. De Meer, and K.S. Trivedi, Queueing Networks and Markov chains, J. Wiley & Sons Inc., New York, 1998.
[2]
{2} S. Dandamudi, "Performance Implications of Task Routing and Task Scheduling Strategies for Multiprocessor Systems", In Proceedings of the IEEE-Euromicro Conference on Massively Parallel Computing Systems, IEEE Computer Society, Ischia, Italy, May 2-6, 1994, pp. 348-353.
[3]
{3} S. Dandamudi, "The effect of Scheduling Discipline on Dynamic Load Sharing in Heterogeneous Distributed Systems", In Proceedings of the 5th International Workshop on Modelling, Analysis, and Simulation of Computer and Telecommunication Systems (MASCOTS '97), IEEE Computer Society, Haifa, Israel, January 12-15, 1997, pp. 17-24.
[4]
{4} H. D. Karatza, "A Simulation-based Performance Analysis of Gang Scheduling in a Distributed System", In Proceedings of 32nd Annual Simulation Symposium, IEEE Computer Society, San Diego, CA, April 11-15, 1999, pp. 26-33.
[5]
{5} H. D. Karatza, "Scheduling Strategies for Multitasking in a Distributed System", In Proceedings of the 33rd Annual Simulation Simposium, IEEE Computer Society, Washington D. C., April 16-20, 2000, pp. 83-90.
[6]
{6} H. D. Karatza, "A Comparative Analysis of Scheduling Policies in a Distributed System using Simulation", International Journal of SIMULATION Systems, Science & Technology, UK Simulation Society, Nottingham, UK, Vol. 1(1- 2), Dec. 2000, pp. 12-20.
[7]
{7} Law, A., and D. Kelton, Simulation Modelling and Analysis, McGraw-Hill, New York, 1991.
[8]
{8} E. Rosti, G. Serazzi, E. Smirni, and M. Squillante, "The Impact of I/O on Program Behavior and Parallel Scheduling", Performance Evaluation Review, ACM, New York, USA, Vol. 26 (1), 1998, pp. 56-65.
[9]
{9} D. Talby, and D. G. Feitelson, "Supporting Priorities and Improved Utilization of the IBM SP2 Scheduler using Slack-based Backfilling", In Proceedings of the 13th International Parallel Processing Symposium and 10th Symposium on Parallel and Distributed Processing, IEEE Computer Society, San Juan, Puerto Rico, April 12-16, 1999, pp. 513-517.
[10]
{10} Y. Zhang, H. Franke, J. Moreira, and A. Sivasubramaniam, "The Impact of Migration on Parallel Job Scheduling for Distributed Systems", In Proceedings of Europar, Europar Munich, Germany, 29 August to 2 September 2000, pp. 242- 251.
[11]
{11} Y. Zhang, and A. Sivasubramaniam, "Scheduling Best-effort and Real-time Pipelined Applications on Timeshared Clusters", In Proceedings of the 13th Annual ACM Symposium on Parallel Algorithms and Architectures, ACM, Crete Island, Greece, July 4-6, 2001, pp. 209-219.
Index Terms
- Performance Analysis of Parallel Job Scheduling in Distributed Systems
Recommendations
Moldable parallel job scheduling using job efficiency: an iterative approach
JSSPP'06: Proceedings of the 12th international conference on Job scheduling strategies for parallel processingCurrently, job schedulers require "rigid" job submissions from users, who must specify a particular number of processors for each parallel job. Most parallel jobs can be run on different processor partition sizes, but there is often a trade-off between ...
Improvements in Parallel Job Scheduling Using Gang Service
ISPAN '99: Proceedings of the 1999 International Symposium on Parallel Architectures, Algorithms and NetworksGang scheduling has been widely used as a practical solution to the dynamic parallel job scheduling problem. Parallel threads of a single job are scheduled for simultaneous execution on a parallel computer even if the job does not fully utilize all ...
Comments
Please enable JavaScript to view thecomments powered by Disqus.Information & Contributors
Information
Published In
March 2003
337 pages
ISBN:0769519113
Copyright © Copyright (c) 2003 Institute of Electrical and Electronics Engineers, Inc. All rights reserved.
Sponsors
Publisher
IEEE Computer Society
United States
Publication History
Published: 30 March 2003
Check for updates
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 595Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 21 Jan 2025
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in