Efficient load balancing techniques for multi-datacenter cloud milieu
- PDF / 2,813,955 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 6 Downloads / 246 Views
ORIGINAL RESEARCH
Efficient load balancing techniques for multi-datacenter cloud milieu Sumanta Chandra Mishra Sharma1,2
•
Amiya Kumar Rath2 • Bivasa Ranjan Parida2
Received: 17 January 2020 / Accepted: 28 September 2020 Bharati Vidyapeeth’s Institute of Computer Applications and Management 2020
Abstract Load balancing is a challenge in the cloud network for proper resource utilization and improvement of makespan. To overcome the problem of load balancing various static and dynamic scheduling methods come into existence. In this paper, we have proposed two multi-datacenter two-phase load adjustment techniques for load balancing and apposite task scheduling. Our proposed methods allocate user jobs to different virtual machines present in different datacenters based on better makespan. Here we have considered communication delay and bandwidth requirement for inter-datacenter task migration. The simulation result specifies that proposed techniques give a significant improvement in makespan and resource utilization when compared with existing algorithms. Keywords Cloud computing Load balancing Multidatacenter approach Communication delay
1 Introduction The spread of technology in the field of computer science introduces a new way to use computational resources. This new technique of easy resource utilization is termed as cloud computing [1], which provides remote computation over the internet. It can also be treated as a technological expansion that emphases on the method we use for designing and developing computational systems and & Sumanta Chandra Mishra Sharma [email protected] 1
Centre of Excellence in Artificial Intelligence, IIT Kharagpur, Kharagpur, India
2
Department of CSE, VSSUT, Burla, India
applications. The cloud milieu proliferates in different geographical locations by providing services on a rental basis to satisfy customer job requirement. To satisfy the need of organizations cloud system provides its services and resources with minimum expenditure. It fulfills the dream of utility computing by providing services on a payas-you-use basis that helps users to use the vast resource with tiny expenditure. In a cloud milieu, the computational resources are distributed and heterogeneous. Each machine in the cloud environment has different configurations, and they differ from one another by their performance, processing power, reliability, etc. Thus the assignment of user jobs to machines are done as per their requirement. As per the business logic in the current scenario, the resources in a cloud system should help the user to get faster and parallel execution of tasks [2, 3]. With that parallel execution, it should ensure that the machines are working correctly with balanced load distribution. If load imbalance occurs between machines in a cloud network, then the scheduler should efficiently schedule the user jobs to bring the system to a balanced position [3]. The essential purpose of task scheduling and load balancing is to maximize the resource utilization [4, 5] and minimize t
Data Loading...
