Dynamic threshold for replicas placement strategy

The data replication is a very important technique for the availability of data in the grids. One of the challenges in data replication is the replicas placement. In this paper, we present our contribution by proposing a replicas placement strategy in a hierarchical grid. Our approach is based on a dynamic threshold, contrary to the other strategies of replicas placement which use a static threshold. In our strategy we show that the threshold depends on several factors such as the size of the data to be replicated, the consumed bandwidth which is explained by the level of the tree related to the grid.