Optimal storage and repair in distributed storage system

Abstract

The distributed storage system maintains the data at multiple nodes in the network to provide faster access to data. In case of failure of one node the data is made available to the users from other nodes and the lost data can be recovered from the surviving nodes. Network coding plays an important role in managing the distributed data. Various coding mechanism like Replication, Erasure Coding, Local Reconstruction, Regeneration, Hybrid and Double Coding have been found beneficial for distributed storage systems. The parameters that make these approaches different from each other are storage requirement, repair bandwidth, node access, fault tolerance etc. This paper proposes a new approach that is capable of repairing the damages with minimum repair bandwidth and occupies less storage compared to other related approaches.