International Journal of Computer Applications (0975 – 8887) Volume 52 – No. 8, August 2012 10 A Balanced Scheduling Algorithm with Fault Tolerance and Task Migration based on Primary Static Mapping (PSM) in Grid Arash Ghorbannia Delavar Department of Computer, Payame Noor University, PO BOX 19395-3697, Tehran, Iran Ali Reza Khalili Boroujeni Department of Computer, Payame Noor University, Tehran, Iran Javad Bayrampoor Department of Computer, Payame Noor University, Tehran, Iran ABSTRACT In this paper we present a balanced scheduling algorithm with considering the fault tolerance and task migration of allocating independent tasks in grid systems. Resource scheduling and its management are great challenges in heterogeneous environment. Hence load balancing is one of the best solutions to achieve the above purposes. The scheduling algorithm which we will present in follow, with taking the fault tolerance, checkpointing method, task migration and priority for mapping independent tasks on heterogeneous computing environment, creates the specific situation to ensure high performance in grid systems. So by implementing these parameters we can achieve more efficient and dependable performance than similar previous algorithms. It will be done with better condition and achieve high performance in computational grids in compare with Min-min algorithm. Finally the experiment and simulated results show that proposed balanced scheduling algorithm performs significantly to ensure high throughput, reduced makespan, reliability and more efficiency in the grid environment. Keywords Grid Computing, Task Scheduling, Heuristic Algorithm, Load Balancing, Fault Tolerance, Task Migration, PSM 1. INTRODUCTION Grid is emerging as a wide scale infrastructure and next generation parallel and distributed computing to aggregates dispersed heterogeneous resources, support source sharing, providing services to fit needs of scientific applications, business, engineering and Commerce [1]. Grid computing environment combination of widely spread computational machines includes of different interconnected machines by interface network to execute different tasks that have diverse computational requirements. A grid involves a variety of resources that are heterogeneous naturally and might span several administrative domains across not narrow geographical distances. Grid computing environment includes of different interconnected machines by interface networks to execute different tasks that have diverse computational requirements. The main purpose of grid systems is optimize using sources and maximizes the efficiency of the system. Managing various resources and task scheduling in grid environment are challenging and indispensable works [2, 3]. Tasks scheduling is a NP- complete problem and finding the absolute optimum solution is too hard. So many heuristic algorithms have been developed to solve this hard problem. The heuristic scheduling can be classified into two categories: on-line mode and batch-mode heuristics. In the on-line mode heuristics, a task is mapped on to a machine as soon as it arrives at the scheduler. In the batch-mode heuristics, tasks are not mapped on to machines as they arrive; instead they are collected into the buffer and then it is scheduled at prescheduled time [4, 5]. Our study is based on the batch-mode heuristics, and presents a batching heuristic scheduling algorithm with consider the fault tolerance and task migration of dedicating independent tasks in grid systems. The scheduling algorithm which we will present in follow executes primary static mapping (PSM) of meta-tasks on the machines in grid systems. Then based on PSM the tasks will be mapped on the machines. The main idea is that if a fault occurs at run time ,or we need to migrate the tasks, the execution will be continued with switching from a processing node to another node, based on PSM (as an optimal target). In this proposed algorithm, the failed machines can be returned to systems to reallocating. By implement the fault tolerance and priority for mapping the tasks in simulated environment we will achieve more efficiently in proposed scheduling algorithm performance, throughput maximization and reduced makespan (measure of the throughput) of the heterogeneous grid computing systems in the grid environment. 2. Related Works Many heuristics algorithms have been designed and developed to solve meta-task optimal scheduling in distributed heterogeneous computing systems. Braun et al. have studied the relative performance of eleven heuristic algorithms for task scheduling in grid computing. They have also provided a simulation basis for researchers to test the algorithms. The simple algorithms proposed by Braun are Opportunistic Load Balancing (OLB), Minimum Execution Time (MET), Minimum Completion Time (MCT), Min-min, Max-min, Duplex, Genetic Algorithm (GA), Simulated Annealing (SA), Genetic Simulation Annealing (GSA), Tabu and A * [6]. The Min-min heuristic begins with the set of all unmapped tasks. Then, the set of minimum completion times is found. Next, the task with the overall minimum completion time is selected and assigned to the corresponding machine. Last, the newly mapped task is removed from unmapped tasks set and the process repeats until all tasks are mapped. Min-min is based on the minimum completion time and considers all unmapped tasks during each mapping decision at a time. Their