Review of Load Balancing in Cloud Computing Suriya Begum Research Scholar. Visvesvaraya Technical University Belgaum, India Dr. Prashanth C.S.R Prof. and Head of Department Dept. of Computer Science and Engineering New Horizon College of Engineering Bangalore, India Abstract— With the exponential rise in the demands of the clients worldwide, a large scale distributed systems have been introduced as a computing environment. Cloud computing has paved a revolutionary path in this direction of distributed environment for accomplishing optimized performance, shortest response time, network resource utilization, and adaptability of service level agreement. Cloud computing has multiple benefits as well as it is also accompanied with certain serious technical loopholes. The proposed paper has focused on one such issue of load balancing. The consequences of inefficient load balancing may lead to detritions of an organization business performance on cloud environment. Hence this paper illustrates various aspects pertaining to domain of cloud computing, its evolution, its generic issues, and particularly to issues related to load balancing. Various techniques adopted in the past research work have been analyzed and the findings were illustrated in this paper. Keywords-; Cloud computing, Grid Computing, Load balancing I. INTRODUCTION loud computing can be illustrated as the operation of heavy computing resources that includes hardware and software package that are delivered to clients as a service over an outsized scale network [1]. The name „cloud‟ has been originated from the employment of a cloud-shaped representation as associate abstraction for the advanced infrastructure that is contained in a very system diagrams. The origination of the term „cloud computing‟ is very vague, however it appears to be derived from the practice of using schematic diagrams of computing and communications systems. The word cloud is employed as a allegory for the big scale network that is supported by the standardized use of a cloud-like form to denote a network on telephone schematics associated later to depict the web in network diagrams as an abstraction of the underlying infrastructure it represents. The cloud image [2] was used to represent the web as early as 1994. The elemental thought of cloud computing essentially dates back to the Nineteen Fifties once there was associate availableness of huge scale mainframe in establishments and companies. Owing to the pricey nature of mainframe, there arise a necessity to search out another answer for allowing the multiple users for accessing and sharing equivalent central processor time thereby truncating the chance of periods of inactivity (also termed as time-sharing) [3]. As computers became additionally prevailing, scientists and technologists explored ways to create large-scale computing power on the market to cater additional users through sharing, experimenting with algorithms to produce the best use of the infrastructure, platform and applications with prioritized access to the computer hardware and potency for the tip users [4]. The high value of those powerful computing systems has forced several prime organizations to require Associate in Nursing initiative to explore higher value effective answer exploitation sharing. The prime organization can embody IBM, GE, National CSS etc who took the initiative to launch and marketed time sharing. With the ascension of net technology and standards, varied merchandise and demands of distributed computing were on high increase. The presence of pervasive high computing network, value effective computing devices, storage devices together with massive scale use of virtualization of hardware, service headed design has paved the trail for top demands in new technology, therefore known as „cloud computing.‟ The domain of cloud computing is still surfaced by many issues which will be discussed in this paper in later section. The prime focus of the paper will be to analyze the research issues in load balancing protocols or understanding its requirement in cloud platform. Since, with the scene of rapid use of cloud computing resources, the demand along with provisioning of the cloud resources has to be effectively design to claim better SLA (Service Level Agreement) with zero downtime claims. Currently, there is a presence of multiple vendors offering cloud services (Amazon, Microsoft, IBM, Google, Salesforce, HP, Oracle, Citrix, EMC etc.) and there are growing numbers of clientele too. Hence, it is quite obvious that catering the massive and dynamic requirements of such exponentially growing clients will become one of the most challenging issues. And to mitigate this issue, an effective load balancing technique should be explored. The proposed paper will introduce a thorough analysis of the evolution of cloud platform right from the origination of the initial distributed computing system. The paper will mainly focus on the research issues of load balancing and will attempt to analyze the prior work done in this field. Section 2 will highlight about the evolution of cloud computing followed by Section 3 for Issues in cloud platform. Section 4 will discuss about the load balancing issues along with the review of the past research work focused on load balancing in cloud platform. Section 5 will discuss about the research issues finally followed by concluding remarks in Section 6 C IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 1, No 2, January 2013 ISSN (Print): 1694-0784 | ISSN (Online): 1694-0814 www.IJCSI.org 343 Copyright (c) 2013 International Journal of Computer Science Issues. All Rights Reserved.