ISSN 2319 - 6629 Volume 4, No.3, April - May 2015
International Journal of Wireless Communications and Networking Technologies Narendrababu Reddy.G etAvailable al., International Journal of Wireless Communications and Network Technologies, 4(3), April - May 2015, 43-47 Online at http://warse.org/pdfs/2015/ijwcnt02432015.pdf
A Weighted Mean Time Selective Scheduling Strategy for Load Balancing of Independent Tasks in Cloud Environment Narendrababu Reddy.G1, Dr. S. Phani Kumar2, 1
2
Asst.Prof.,GNITS, Research Scholar, GITAM,Hyderabad,gnbreddy25@gmail.com Prof. & HOD, CSE Department, GITAM University, Hyderabad, phanikumar.s@gitam.edu
ABSTRACT their requirements without regard to where the services are hosted or how they are delivered. Cloud computing [2] is an entirely internet based approach where all the applications and files are hosted on a cloud which consists of thousands of computers interlinked together in a complex manner. Cloud computing incorporate concepts of parallel and distributed computing to provide shared resources: hard ware, software and information to computers are other devices on demand. The end users can use these resources over a network on-demand basis as a “pay- per- use” model. The customer is interested in reducing the overall execution time of tasks on the machines. This leads to problems in scheduling of customer tasks within available resources. In cloud platforms, load balancing (or resource allocation) takes place at two levels. First, when an application is uploaded to the cloud, the load balancer assigns the required instances to physical computers. Second, when an application receives multiple incoming requests, these requests should be each assigned to a specific application to balance the computational load across a set of instances of the same application. Load balancing must take into account the consideration of all kinds of cloud users. It must fulfill the requirements of all its users. The cloud users expect their jobs to be completed in fastest possible manner with high availability of resources. On the other hand the cloud provider invests such a huge capital with the aim of maximum utilization of all the installed resources in efficient and effective manner. Both cloud users and cloud providers except maximum throughput and improved performance for the money they invest. The main objective of the load balancing methods is to speed up the execution of applications on resources whose workload varies at run time in unpredictable way[13]. Motivated by this problem, we propose a new load balancing algorithm for cloud systems to allocate resources to tasks using min-min or min-max algorithm and then scheduling them on either space shared or time shared basis. The rest of this paper is organized as follows. The second section describes the related works on existing load balancing techniques. Section three describes the proposed load balancing algorithm. Section four presents experimental results along with performance evaluation of the algorithm in comparison with existing algorithms followed by concluding remarks in section five.
Cloud computing nowadays becomes quite popular among a community of cloud users by offering a variety of resources. In cloud computing environments, resources and infrastructure are provided as a service over internet on demand. These resources need to be provisioned to the end users in most efficient manner to satisfy the SLAs. Load balancing of independent tasks in cloud computing environment is a crucial issue of resources allocation to the user requirements. In this paper we are presenting a new heuristic scheduling strategy for allocation of cloud resources to end users tasks on demand basis, which aims to achieve well balanced load across virtual machines for maximizing the throughput. This algorithm uses certain heuristics to select between two algorithms so that overall make-span of tasks on the machines is minimized. We evaluate our provisioning heuristics by comparing with existing load balancing and scheduling algorithms. Our approach illustrates that overall make-span of tasks on given set of VMs minimizes significantly in different scenarios. Key Words— Cloud computing, Load balancing, Make-span, Min-Min, Max-Min. 1. INTRODUCTION With the rapid development of processing and storage technologies and evolution of internet, computing resources have become cheaper, more powerful and more ubiquitously available than ever before. This technological trend has enabled the realization of a new computing model called Cloud Computing, in which resources are provided as general utilities that can be leased and released by users through the internet in an on-demand fashion. According to Rajkumar Buyya et al.[1] “A Cloud is a type of parallel and distributed system consisting of a collection of inter-connected and virtualized computers that are dynamically provisioned and presented as one or more unified computing resource(s) based on service-level agreements established through negotiation between the service provider and consumers.” Computing is being transformed to a model consisting of services that are commoditized and delivered in a manner similar to traditional utilities such as water, electricity, gas and telephone. In such a model, users access services based on 43