Dynamic Load Balancing Algorithm in the Global-Local Order

  IJCTT-book-cover
 
International Journal of ComputerTrends and Technology (IJCTT)          
 
© 2014 by IJCTT Journal
Volume-7 Number-1                          
Year of Publication : 2014
Authors : Hojiev Sardor Qurbonboyevich , Tae-Young Choe.
DOI :  10.14445/22312803/IJCTT-V7P115

citation

Hojiev Sardor Qurbonboyevich , Tae-Young Choe. Article: Dynamic Load Balancing Algorithm in the Global-Local Order. International Journal of Computer Trends and Technology (IJCTT) 7(1):26-34, January 2014. Published by Seventh Sense Research Group.


Abstract
-Most traditional dynamic load balancing schemes for hierarchical environments have applied local load balancing first and have expands it to the global load balancing. The major problem of the approach is that unnecessary task immigrations can occur, which degrade the system performance. Carefully designed global load balancing scheme eliminates the unnecessary task immigrations. We propose a dynamic load balancing scheme that balances global level first followed by local level. Two thresholds that include communication overheads are applied to the load balancing scheme. Experiments show that the proposed scheme 99.7% and 87.2% of average response time than traditional local-first-global-later load balancing scheme, in the case of a uniformly distributed workload and a single hot spot workload, respectively.

References

[1] Ian Foster, Carl Kesselman, Jeffrey M Nick, and Steven Tuecke, “Grid services for distributed system integration”, Computer, 35(6), pp. 37-46, 2002.
[2] Ian Foster, Yong Zhao, Ioan Raicu, and Shiyong Lu, “Cloud computing and grid computing 360-degree compared”, in Grid Computing Environments Workshop, 2008. GCE`08. IEEE, 2008,pp. 1-10.
[3] Thomas L. Casavant and Jon G. Kuhl, “A taxonomy of scheduling in general-purpose distributed computing systems”, Software Engineering, IEEE Transactions on, 14(2), pp. 141-154,1988.
[4] Mohammed JaveedZaki, Wei Li, and SrinivasanParthasarathy, “Customized dynamic load balancing for a network of workstations”, in High Performance Distributed Computing, 1996., Proceedings of 5th IEEE International Symposium on. IEEE, 1996, pp. 282-291.
[5] BelabbasYagoubi and YahyaSlimani, “Dynamic load balancing strategy for grid computing”, Transactions on Engineering, Computing and Technology, 13, pp. 260-265, 2006.
[6] BelabbasYagoubi and Meriem Meddeber, “Distributed load balancing model for grid computing”, Revue ARIMA, 12, pp. 43-60, 2010.
[7] Said Fathy El-Zoghdy, “A load balancing policy for heterogeneous computational grids”, International Journal of Advanced computer Science and Applications, 2(5), pp. 93-100, 2011.
[8] MalarvizhiNandagopal and Rhymend V Uthariaraj, “Hierarchical status information exchange scheduling and load balancing for computational grid environments”, IJCSNS International Journal of Computer Science and Network Security, 10(2), pp. 177-185, 2010.
[9] MinakshiTripathy and CR Tripathy, “Dynamic load balancing with work stealing for distributed shared memory clusters”, in Industrial Electronics, Control & Robotics (IECR), 2010 International Conference on. IEEE, 2010, pp. 43-47.
[10] SeyedRasoolMoosavi-Nejad, S. S. Mortazavi, and BijanVosoughiVahdat, “Fuzzy based design and tuning of distributed systems load balancing controller”, in 5th Symposium on Advances in Science & Technology (SASTech), Mashhad, Iran, May 2011.
[11] RajkumarBuyya and ManzurMurshed, “Gridsim: A toolkit for the modeling and simulation of distributed resource management and scheduling for grid computing”, Concurrency and Computation: Practice and Experience, 14(13-15), pp. 1175-1220, 2002.
[12] HojievSardorQurbonboyevich and Tae-Young Choe, “Two-level dynamic load balancing algorithm using load thresholds and pairwise immigration”, International Journal on Computer Science and Engineering (IJCSE), 5(4), pp. 211-220, April 2013.

Keywords- Dynamic Load Balancing, Grid computing, GridSim, Hierarchical distributed System, Multiple Threshold.