Deepika Sharma al. International Journal of Recent Research Aspects ISSN: 2349-7688, Vol. 1, Issue 2, Sept. 2014, pp. 5-9 © 2014 IJRRA All Rights Reserved page - 5- Evaluating Heuristic based Load Balancing Algorithm through Ant Colony Optimization Deepika Sharma 1 , Surjeet Dalal 2 , Kamal Kumar Sharma 3 1 Student, M. Tech, ESEAR, Ambala 2 Assistant Professor, Dept. of CSE, E-Max group of Institutions, Ambala 3 Professor, Dept. of ECE, E-Max group of Institutions, Ambala Abstract— Grid is an infrastructure that involves the integrated and collaborative use of computers, networks, databases and scientific instruments owned and managed by multiple organizations. Computational grid has been considered as the best paradigm for handling large scale distributed system having geographically allocated resources. Load balancing algorithms are important in the research of network applications. In this paper we present an algorithm which reduces the average execution time and cost of the tasks. This method considers both cost and time constraints. This algorithm provides services like resource discovery. For evaluation purpose a comparison of execution times and cost of proposed algorithm and the other similar algorithm is also provided in this paper. Results support the proposed approach. Keywords— Grid, load balancing, heuristic Load balancing algorithm, Gridsim, computation time.. I. INTRODUCTION Poster and Kesselman introduced grid computing as a way to utilize the geographically distributed available idle workstation‟s computation power to remote grid users for the execution of their computation requiring jobs or tasks or processes. Grid Computing enables sharing, selection, aggregation of geographically distributed resources dynamically at run time depending on their accessibility, ability and users Quality of Service requirements The user essentially interacts with a resource broker that hides the complexities of Grid computing [4,5]. The broker discovers resources that the user can access using information services, negotiates for access costs using trading services, maps tasks to resources (scheduling), stages the application and data for processing (deployment), starts job execution, and finally gathers the results. It is also responsible for monitoring and tracking application execution progress along with adapting to the changes in Grid runtime environment conditions and resource failures as the grid resource utilization is gaining significance various scheduling methods or load balancing across the grid is required in order to improve the performance of the grid system. While balancing the load, certain type of information such as the number of jobs waiting in queue, job arrival rate, CPU processing rate, and so forth at each processor, as well as at neighboring processors, may be exchanged among the processors for improving the overall performance. Based on the information that can be used, Scheduling is mainly classified into two types static and dynamic scheduling or adaptive [7,10].Static approach assumes that a prior information about all the characteristics of the jobs, the computing devices and the communication network are known and provided. Load balancing decisions are made deterministically or probabilistically at compile time and remain constant during runtime. Dynamic load balancing algorithm attempts to use the runtime state information to make more informative decision in sharing the system load. However, dynamic scheme is used in modern load balancing method a lot due to their robustness and flexibility. Dynamic algorithm is characterized by parameters such as i) Centralized vs. Decentralized. An algorithm is centralized if the parameters necessary for making the load balancing decision are collected at, and used by, a single device. In decentralized approach all the devices are involved in making the load balancing decision. Scalability and fault tolerance are the advantages of decentralized algorithms.. ii) Cooperative vs. Non-cooperative. An algorithm is said to be cooperative if the distributed components that constitute the system cooperate in the decision-making process. Otherwise, it is non-cooperative.and iii) Adaptive vs. Non-adaptive. If the parameters of the algorithm can change when the algorithm is being run, the algorithm is said to be adaptive (to the changes in the environment in which it is running). Otherwise, it is non-adaptive. The characteristic of grid makes resource scheduling even more challenging in the computational grid environment. In this paper we propose a new load balancing algorithm which is heuristic in nature. Unlike previous algorithms which considered the system load of grid nodes or the completion time for a job but don‟t take into account job personal resource requirements (such as cost , QOS of a node) . In this paper we present an algorithm which considers the job size and is mainly focused on formulating a decentralized heuristic based load balancing algorithm for resource scheduling. The rest of the paper is organised as follows: Section II gives a detailed Literature review. Section III deals with the system model of the Grid. Section IV will describe the proposed algorithm concept and design. Section V will show the simulation experiment and results, and finally Section VI will conclude the whole paper.