Abstract: “In grid computing, resources are used outside the boundary of organizations and it becomes increasingly difficult to guarantee that resources being used are not malicious. Also, resources may enter and leave the grid at any time. So, fault tolerance is a crucial issue in grid computing. Fault tolerance can enhance grid throughput, utilization, response time and more economic profits. All mechanisms proposed to deal with fault-tolerant issues in grids are classified into: job replication and job checkpointing techniques. These techniques are used according to the requirements of the computational grid and the type of environment, resources and virtual organizations it is supposed to work with. Each has its own advantages and disadvantages which forms the subject matter of this paper.”

Keywords: Fault tolerance, Grid computing, Checkpointing, Job replication.