Queuing Analysis of Cloud Load Balancing Algorithms

Queuing Analysis of Cloud Load Balancing Algorithms

Santosh Kumar Majhi (VSS University of Technology, Department of Computer Science and Engineering, Burla, India), Shankho Subhra Pal (VSS University of Technology, Department of Computer Science and Engineering, Burla, India), Shweta Bhuyan (VSS University of Technology, Department of Computer Science and Engineering, Burla, India) and Sunil Kumar Dhal (Sri Sri University, Faculty of Management Studies, Cuttack, India)
Copyright: © 2018 |Pages: 18
DOI: 10.4018/IJKBO.2018010104
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The emergence of cloud-computing and the apparent shift to this new paradigm has led to the creation of data centres that consist of hundreds of thousands of servers. The Cloud is a distributed system that helps share data and provides resources to the users. The data and the distributed resources are stored in the open environment. This paper presents a model of cloud load balancing using queuing and probability theory. A queuing cloud model is discussed with load balancing perspective. We present analysis for two servers and then extended it to n server. In addition, an optimal strategy is modelled for cloud load balancing. The analytical results are verified through numeric simulation.
Article Preview

Load Balancing Algorithms

Cloud computing is the recent paradigm of large scale distributed computing. Optimal resource utilisation is the necessary prerequisite of parallel and distributed systems. Load balancing is the method that distributes the workload among servers, data centres, hard - drives or other computing resources which are regarded as ‘nodes’ in the given environment such that a node is neither over - whelmed nor under - utilised. The unpredictability in the cloud environment is caused due to the dynamic behaviour of the cloud which makes load balancing a crucial concern in cloud computing. Load Balancing is nothing but an optimisation problem and an efficient load balancing algorithm is the one that adapts itself to the stochastic environment and the diverse tasks. However, allocating the request uniformly across the nodes is considered to be an NP - complete problem (Baca, 1989).

Since scheduling is NP – Hard optimisation problem, so (Babu & Krishna, 2013) have used a technique inspired by Honey Bee Behaviour, to schedule the load among different VM (Virtual Machines).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 8: 4 Issues (2018): 1 Released, 3 Forthcoming
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing