A Cognitive Analysis of Load Balancing and job migration Technique in Grid Neeraj Rathore Dr. Inderveer Chana Department of CSE Department of CSE Thapar University Thapar University Patiala, Punjab, India Patiala, Punjab, India neerajrathore37@gmail.com inderveer@thapar.edu Abstract- To improve the global throughput of grid environments, workloads have to be evenly balanced among the available resources. However for computational grid we must address main new issues. Like heterogeneous autonomy, dynamicity and so forth. A framework consisting of distributed dynamic load balancing algorithm in perspective to minimize the average response time of tasks. The main goal is to prevent, if possible, the condition where some processors are overloaded with a set of tasks while others are lightly loaded or even idle. This is due to the characteristics of Grid computing and the complex nature of the problem itself. Load balancing and job migration allocate processes to interconnected workstations on a network to better take advantage of available resources. Job migration in a distributed computer system can be performed for performance enhancement and we call this activity\load balancing with job migration. The intractability of this load balancing model suggests obtaining approximate solutions. In this paper, we Connitive analysis of differnt approaches of load balancing and job migration into distributed system. By using a good technique approximate solutions can be obtained. Our contribution in this paper is to find out the loophole in all the previous paper related in load balancing and job migration and work upon on those gaps in my research work and it is beneficial for all those researches who want to make more enhance and efficient load balancing techniques in near future. Key Words: Grid, Distributed system, grid computing, load balancing, Job Migration and workload. I. INTRODUCTION Grid is a type of distributed system that supports the sharing and coordinated use of geographically distributed and multi- owner resources independently from their physical type and location in dynamic virtual organizations that share the same goal of solving large-scale applications [12]. A computational grid is a hardware and software infrastructure that provides dependable, consistent, pervasive, and inexpensive access to high-end computational capabilities. Grid computing is concerned with “coordinated resource sharing and problem solving in dynamic, multi- institutional virtual organizations” [13]. The key concept is the ability to negotiate resource-sharing arrangements among a set of participating parties (providers and consumers) and then to use the resulting resource pool for some purpose [14]. Distributed systems are collections of autonomous processing nodes connected by a communication network. Through the communication network, the resources of the system can be shared by users at different locations. However, a fundamental problem arises in making effective use of the total computing power of a distributed computing system. It is often the case that a certain node has very few tasks to handle at a given time, while another node has many. It is desirable to spread the total workload of the distributed system over all of its nodes. This avoids under-utilization of power, and decreases response times for work introduced at more heavily loaded sites. This form of computing power sharing for improving the performance of a distributed system by redistributing the workload among the available nodes is commonly called “load balancing”. The purpose of load balancing is to improve the performance of a system by redistributing the workload among nodes, thus increasing the processing capacity of the system [1] [2]. Resource management and load balancing are key grid services, where issues of local balancing represent a common concern for the most grid infrastructure developers [11] [18]. The computing power of any system does not increase proportionally with the number of resources involved. Therefore there is a need to continuously monitor that some resources do not become overloaded and some others stay idle [15]. The essential objective of a load balancing consists primarily in optimizing the average response time of applications, which often means maintaining the workload proportionally equivalent on the entire resources of a system [17]. Grid Load Balancing is based on the idea of migration of excess load from heavily loaded node to lightly loaded ones. The problem starts with to determine when to migrate a process or task. This solution is typically based on local load situation [19]. Load balancing algorithms vary in their complexity where complexity is measured by the amount of communication used to approximate the least loaded node. The most significant parameter of the system is the cost of transferring a job from one node to another [20]. Efficient load balancing across the Grid is required for improving performance of the system. The overloaded grid resources can be balanced by migrating jobs to the buddy processors, i.e. a set of processors to which a processor is directly connected. Job migration means re-allocation of jobs from one system to another system or in other words job migration is the process of moving a normal, checkpointable or rerunable job from one host to another [28]. It can be initiated manually or automatically. Job migration can be a solution of the load balancing problem because it is required to balance the load whenever load imbalance problems come in distributed environment. Grid is such a system, where environmental conditions are subject to unpredictable changes: system or network failures, system performance degradation, addition or deletion of new machines, variance cost of resources, etc. In such conditions, job migration is the only efficient way to guarantee that the submitted jobs are completed as per user requirements [27]. 77 978-1-4673-0126-8/11/$26.00 c 2011 IEEE