A new time optimizing probabilistic load balancing algorithm in grid computing

The computing grid is a distributed parallel processing system that share and choosing resources dynamically and provide need of user operation power, cost and quality .grid management resources does as a diagnostic and assigning resources scheduling and resource monitoring in grid . Scheduling process directs tasks to suitable resources. It must take place some how that load work distributed equally on resources to get the maximum interest out of existed resource, establishing load balancing is one the important performance factors in grid resource management efficiency. in this paper , loading indexes and new resource conditions in accordance with synchronous neighbourhood was suggested and also for resource allocation ,a model in accordance with tree and probabilistic scheduling algorithm with load balancing purpose was suggested , that in this algorithm workclass, cost, deadline and herd behaviour have considered. Probabilistic algorithm chooses the resources that have better past and least completion time And leave the duties to it, in case of execution or non-execution on the resource the source will give a reward or punishment. The main purpose of this algorithm is establishing load balancing and reducing the response time and task failure percentage.

[1]  Yahya Slimani,et al.  Dynamic Load Balancing Strategy for Grid Computing , 2006 .

[2]  Alba Cristina Magalhaes Alves de Melo,et al.  Using a classifier system to improve dynamic load balancing , 2001, Proceedings International Conference on Parallel Processing Workshops.

[3]  Ravi Mirchandaney,et al.  Using Stochastic Learning Automata for Job Scheduling in Distributed Processing Systems , 1986, J. Parallel Distributed Comput..

[4]  Ian Foster,et al.  What is the Grid , 2002 .

[5]  Sajal K. Das,et al.  A de-centralized scheduling and load balancing algorithm for heterogeneous grid environments , 2002, Proceedings. International Conference on Parallel Processing Workshop.

[6]  Michael Mitzenmacher,et al.  How Useful Is Old Information? , 2000, IEEE Trans. Parallel Distributed Syst..

[7]  Bharadwaj Veeravalli,et al.  On the Design of Adaptive and Decentralized Load Balancing Algorithms with Load Estimation for Computational Grid Environments , 2007, IEEE Transactions on Parallel and Distributed Systems.

[8]  Jie Pan,et al.  Introduction to Grid Computing , 2009 .

[9]  Debra A. Hensgen,et al.  The relative performance of various mapping algorithms is independent of sizable variances in run-time predictions , 1998, Proceedings Seventh Heterogeneous Computing Workshop (HCW'98).

[10]  N. Ranganathan,et al.  Multiple cost optimization for task assignment in heterogeneous computing systems using learning automata , 1999, Proceedings. Eighth Heterogeneous Computing Workshop (HCW'99).

[11]  R. F. Freund,et al.  Scheduling resources in multi-user, heterogeneous, computing environments with SmartNet , 1998, Proceedings Seventh Heterogeneous Computing Workshop (HCW'98).