Dynamic Self-Adaptive Replica Location Method in Data Grids

Data replication is a general mechanism to improve performance and availabilityfor distributed applications. However, it is a challenging problem to find thephysical locations of one or more replicas of desired data efficiently in large-scale data grid systems. To solve the problem, a new dynamic self-adaptive distributed replica location method (DSRL) is proposed. In DSRL each data element has a home node, which maintains the index of the location information of replicas. Home nodes are used to accelerate the process of locating multiple replicas of the same data element. Meanwhile DSRL employs local location nodes which maintain the local replica information of data elements to support local query for replicas. A dynamic balancing technique that can adapt to the joining or departing of home nodes is proposed to spread global replica location information evenly on location nodes. The correctness and properties of DSRL are presented and proved. Analysis and experiments show that DSRL can achieve good scalability, low latency, reliability, adaptability and ease of implementation.