A Dynamic Different Threshold Strategy to Data Replication in Data Grids

Abstract

In the context of data grid technology, data replication is mostly used to reduce access latency and bandwidth consumption. Two of well-known replication strategies are 􀀀 -

and , which work well for different kinds of access patterns individually. Later, Chang propose the static different threshold ( ) strategy, which can work for any kind of access patterns. However, among a large number of different data files, there may exist some hot data files which have been most requested. To reduce the number of requests for the hot files, we should replicate those hot files as soon as possible. Therefore, in this paper, we propose the dynamic DT strategy. In the dynamic DT strategy, data replication of hot files occurs earlier than others because the thresholds of hot files are decreased earlier than thresholds of the normal files. From our simulation results, we have shown that the performance of the dynamic DT strategy is better than the previous strategies.