Replica Consistency Maintenance And Selection In Data Grid

In the Internet environment, there exist massive data resources scattered at distributed locations with different types. The management of massive data becomes a challenging problem because of its high heterogeneity, decentralization and complexity of sharing. Data Grid integrates with high-performance computing facilities and massive storage equipments. It realizes many data management functionalities, such as data storage, data access, data transport, and replica management. It is regarded as a novel infrastructure with justice, self-adaptability and inter-activity for massive data management and sharing.Replica Management was introduced in data grid to enhancing data availability, reduce network traffic, and improve data access performance. It becomes one of the hot spot issues in literature to build an appropriate replica management mechanism to enhance data availability and data access performance according to data grid characteristics.Based on the development requirement of data access mode in data grid, we build a replica consistence maintenance model after we analyze the problems that could exist in process of current replica consistency maintaining, break the problems into local replica consistency, update conflicts detection and resolution, and update propaganda which means a lot to consistency performance and resolve them separately. The experiment and simulation on OptorSim shows that update propaganda algorithm we designed could bring efficient consistency performance.The changing of data access mode also leads to that of metrics on replica selection. Due to replica update, replicas exist in data grid could be different, resulting in replica access requirement changing. So we study replica selection problems, and propose a price model that could be used in replica selecting. And then we implement this replica selection algorithm based on replica consistency implementation, simulation shows that this algorithm comes up to meet our expectations.