Replication Strategies for Data Grids

The distributed and dynamic nature of grids present a number of challenging problems distinguished from conventional distributed computing by its focus on large-scale resource sharing and high-performance computing. Grid technology facilitates data sharing across many organizations in different geographical locations. Grid enabled applications need to manage and process large amount of data distributed across multiple grid nodes and stored in heterogeneous databases. Data replication is an excellent technique to move and cache data close to users. Replication reduces access latency and bandwidth consumption. It also facilitates load balancing and improves reliability by creating multiple data copies. Selecting a site for the placement of replica is attributed to various factors such as number of requests for a particular file, bandwidth, read/write statistics and location of the resources. In this paper, three replication strategies for dynamic data grids namely best replica site, cost effective site and topology based replication have been proposed and compared with no replication and existing methodologies.