Abstract
Desktop grid is based on desktop computers owned and volunteered by individual users. Volunteer nodes donate their spare capacity like CPU cycles, memory and any other shared resources during their free time for public execution and withdraw from the public execution during their busy time. This is due to their high priority private jobs initiated by the node owner. Therefore, the volunteer nodes may join or leave the desktop grid at any instant of time because the volunteer nodes are not dedicated to public execution. Hence, the volunteer nodes availability time for public execution varies from one node to another node. In this paper, we consider the volunteer interferences as volunteer failures and duration of the interference (time to rejoin for public execution) as volunteer repair time. When a node fails during the execution of a job, the job might need to be resumed/restarted based on the fault tolerant capability of the node. This causes slowdown in the execution of the jobs at some nodes than the other nodes having the similar processing capability. This situation often leads to increased computational demands at some nodes. The desire to meet the increased computational demand at each node has influenced the interest on job scheduling policies that migrates local jobs for remote execution for the better turnaround time. In this paper we propose a prediction based job scheduling strategy (PJSS) that makes use of neural network load predictions combined with the node reliability parameters for making job scheduling decisions. The performance of the proposed method is compared against no-migration (NM) and resource exclusion (RE) algorithms. The simulation results show that the average turnaround time per job for the proposed method has got considerable improvement over no-migration and resource exclusion algorithms.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Hamscher, V., Schwiegelshohn, U., Streit, A., Yahyapour, R.: Evaluation of Job-Scheduling Strategies for Grid Computing. In: Buyya, R., Baker, M. (eds.) GRID 2000. LNCS, vol. 1971, pp. 191–202. Springer, Heidelberg (2000)
Yagoubi, B., Slimani, Y.: Task Load Balancing Strategy for Grid Computing. Journal of Computer Science 3(3), 186–194 (2007)
Fedak, G., Germain, C., Neri, V., Cappello, F.: XtremWeb: A Generic Global Computing System. In: CCGRID 2001 (2001)
Casanova, H., Legrand, A., Zagorodnov, D., Berman, F.: Heuristics for Scheduling Prameter Sweep Applications in GRID Environments. In: HCW 2000 (2000)
Wolski, R., Spring, N., Hayes, J.: The Network Weather Service: A Distributed Resource Performance Forecasting Service for Metacomputing. In: Future Generation Computing Systems (1999)
He, X., Sun, X.-H., Laszewski, G.V.: QoS guided Min-min heuristic for grid task scheduling. Journal of Computer Science and Technology 18, 442–451 (2003)
Estiminani, K., Naghibzadeh, M.: A Min-min and Max-min selective algorithm for Grid task scheduling. In: The Third IEEE/IFIP International Conference on Internet, Uzbekistan (2007)
Dinda, P.: Online prediction of the running time of tasks. Cluster Computing 5(3), 225–236 (2002)
Entropia, Inc., http://www.entropia.com
Yu, J., Buyya, R.: A taxonomy of scientific workflow systems for grid computing. SIGMOD Record 34(3) (September 2005)
Zhou, D., Lo, V.: Wave Scheduler: Scheduling for Faster Turnaround Time in Peerbased Desktop Grid Systems. Presented at 11th Workshop on Job Scheduling Strategies for Parallel Processing In Conjunction with ICS 2005. The Cambridge Marriott-Kendall Square, Cambridge (2005)
Kondo, D., Kindarji, B., Fedak, G., Cappello, F.: Towards Soft Real-Time Applications on Enterprise Desktop Grids. In: Sixth IEEE International Symposium on Cluster Computing and the Grid (CCGRID 2006), pp. 65–72. IEEECS Press (May 2006)
Choi, S.J., Baik, M.S., Hwang, C.S., Gil, J.M., Yu, H.C.: Volunteer Availability based Fault Tolerant Scheduling Mechanism in Desktop Grid Computing Environment. In: The 3th IEEE International Symposium on Network Computing and Applications, Workshop Ouun Adaptive Grid Computing (NCA-AGC 2004), pp. 476–483 (August 2004)
Choi, S.J., Baik, M.S., Gil, J.M., Jung, S.Y., Hwang, C.S.: Adaptive Group Scheduling Mechanism using Mobile Agents in Peer-to-Peer Grid Computing Environment. Applied Intelligence, Special Issue on Agent-based Grid Computing 25(2), 199–221 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Naseera, S., Madhu Murthy, K.V. (2013). Prediction Based Job Scheduling Strategy for a Volunteer Desktop Grid. In: Unnikrishnan, S., Surve, S., Bhoir, D. (eds) Advances in Computing, Communication, and Control. ICAC3 2013. Communications in Computer and Information Science, vol 361. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-36321-4_3
Download citation
DOI: https://doi.org/10.1007/978-3-642-36321-4_3
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-36320-7
Online ISBN: 978-3-642-36321-4
eBook Packages: Computer ScienceComputer Science (R0)