Open AccessEfficient Load Balancing and Service Broker Policy Using Local and Global Representatives in Cloud Computing
Author(s) -
Amrita Jyoti,
Dr Shiv Singh Saragdevot,
Deepak Chandra
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.j9445.0881019
Subject(s) - computer science , cloud computing , load balancing (electrical power) , distributed computing , virtual machine , scheduling (production processes) , task (project management) , service level agreement , utility computing , cloud testing , service (business) , operating system , cloud computing security , operations management , geometry , mathematics , management , economy , economics , grid
Cloud computing is a technology in the field of computing which offer services to the customer from anywhere at any time [1]. In the cloud, resources are shared all around the work for quick servicing to the customer. The aggregation of two terms is referred as cloud computing. The term “cloud” is a pool of different resources offers services to the end customers and “computing” is done based on the Service Level Agreement (SLA) to make the resources efficiently to the customers. Load balancing is an important challenge in the environment of the cloud to increase the utilization of resources [3]. Here we proposed an algorithm which is based on load balancing and service broker policy. We user two representative thin the proposed approach local representative and global representative Local user representative is used to predict the parameters of user task and based on priority it allocate the task to the Virtual Machine (VM). Then for scheduling the task and provide the services to the users based on the available cloud brokers global user representative used Dynamic Optimal Load-Aware Service Broker (DOLASB).we used two scenario with different no. of user requests , in these scenario result of our proposed method is better as compared with the other existing methods in terms s of Execution Time, Makespan, Waiting Time, Energy Efficiency and Throughput.