HABCO: A Hybrid Algorithm to Improve Load Balancing in Cloud Computing

Volume 4 Issue 1 January - June 2017

Research Paper

HABCO: A Hybrid Algorithm to Improve Load Balancing in Cloud Computing

K. Tamilarasi*, V. Ramya**

** M.Phil Scholar, Department of Computer Science, Government Arts College for Men, Nandanam, Chennai. India

*** Assistant Professor, Department of Computer Science, Government Arts College for Men, Nandanam, Chennai, India.

Tamilarasi. K., and Ramya, V. (2017). HABCO: A Hybrid Algorithm to Improve Load Balancing In Cloud Computing. i-manager's Journal on Cloud Computing, 4(1), 15-22. https://doi.org/10.26634/jcc.4.1.13754

Abstract

Cloud Computing is known to be a well emerged architecture in recent times that tends to satisfy all the aspirations of entrepreneurs and researchers on a variety of fields. In the modern era we find large volume of data in different forms to be stored on cloud and multiple requests are triggered for resources, thus increasing load on servers sometimes unable to provide all the available resources on required time. This piece of research work designs a hybrid algorithm that is framed by combining the characteristics of two load balancing algorithms. Hybrid Ant Bee Colony optimization algorithm is a new phase in the field of load balancing on cloud computing that implements the characteristics real ants and bees. The searching of food source of an ant and searching of honey hive of bee resembles the searching of nodes in task scheduling. The algorithm designed implements the Ant colony optimization concept that corresponds updating pheromone table. Frequently updating the Pheromone table improves the performance of load balancing, the magnitude of pheromone deposited in the path of food source and also the fitness value calculation of the bees helps in identifying the best hive of bees. The works draw a different perspective by calculating the waiting time for a node to get allocated by a new task. It assures the improvement in performance of the system when HABCO is implemented as the node undergoes less waiting time.