A realistic distributed storage system: the rack model

Abstract: In a realistic distributed storage environment, storage nodes are usually placed in racks, a metallic support designed to accommodate electronic equipment. It is known that the communication (bandwidth) cost between nodes which are in the same rack is much lower than between nodes which are in different racks. In this paper, a new model, where the storage nodes are placed in two racks, is proposed and analyzed. Moreover, the two-rack model is generalized to any number of racks. In this model, the storage nodes have different repair costs depending on the rack where they are placed. A threshold function, which minimizes the amount of stored data per node and the bandwidth needed to regenerate a failed node, is shown. This threshold function generalizes the ones given for previous distributed storage models. The tradeoff curve obtained from this threshold function is compared with the ones obtained from the previous models, and it is shown that this new model outperforms the previous ones in terms of repair cost.