ZAHRA MIROEI 1 AND MARJAN KUCHAKI RAFSANJANI 2 1. Department of Computer, Science and Research Branch, Islamic Azad University, Kerman, Iran
2. Department of Computer Science, Shahid Bahonar University of Kerman, Kerman, Iran, e-mail: kuchaki@uk.ac.ir
Today, scientific research and commercial applications generate large amounts of data. In such circumstances, storage and management of these data volume centrally is difficult or even impossible. Data Grid is a good solution for data management, storage resource management and execution of applications. In applications related to data, re-quired data transmission time is the main cause of delay in task execution. In order to decrease it, one method is to put the required data in places near task execution. On the other hand, one of the tasks of data grid is to share data and increase data availability and reliability. In order to reach these goals, one suitable solution is to replicate data in differ-ent parts of the system. In this method, not only required data for task execution are repli-cated near it and file transmission time and task execution time are decreased, also data availability is increased, system reliability is increased, fault tolerance is increased, system efficiency and scalability are increased. Benefits of data replication have led to many researchers to study in this field and propose many algorithms. In this paper the research-es done in this field are studied and compared.
