Affiliation:
1. Gandhinagar Institute of Technology, India
Abstract
Grid computing has emerged as a striking computing concept due to the availability of high-speed wide area networks and low-cost computational capital. This represents a novel load balancing algorithm for grid systems by allowing for node allure. The set of partners and neighbours are created for each node using the prestige of the job. For each job, in the gird, the given algorithm uses the popularity of other nodes in the grid to form k number of partners and p number of neighbours. The methods for constructing neighbours and partners are existing. A new job arriving to a node is immediately dispersed to the originating node or to its partner nodes. The load adjustment is approved incessantly, and reciprocal information administration is used to minimize the communication overhead in the elaborated load balancing algorithm. The given algorithm is self-motivated, sender-initiated, and decentralized.
Reference13 articles.
1. Ashok, A. S., & D., P. (2012). Grid Computing: various job scheduling strategies, emerging trends in computer science and information technology. In Proceedings of the International Conference on Emerging Trends in Computer Science and Information Technology. ETCSIT.
2. Benmohammed-Mahieddine, K. (n.d.). An Evaluation of Load Balancing Algorithms for Distributed Systems. Academic Press.
3. An Adaptive Scoring Job Scheduling algorithm for grid computing.;R.-S.Chang;Information Sciences,2012
4. El-Zogdhy, S., Kameda, H., & Li, J. (2006). Numerical studies on a paradox for noncooperative static load balancing in distributed computer systems. Computers and Operations Research, 345-355.
5. Diffusive load balancing schemes on heterogeneous networks