Abstract:
Peer-to-Peer (P2P) based distributed storage systems have gain much popularity in recent years. These systems rely greatly on the data redundancy to be robust under network dynamics, i.e., the dynamics of peer entering and departing the network. Hence, it is important to implement mechanisms for maintaining a certain level of data redundancy at all times in the network. One such mechanism is called distributed data replenishment which attempts to repair data due to a peer failure or departure in a distributed manner. Such distributed data replenishment schemes make use of the well-known Repetition code, the Reed Solomon code, or recently the Random Linear Network Coding techniques. However, these schemes do not consider bandwidth associated with peers during the replenishment. In this thesis we explore techniques for fast replenishment by taking into consideration the bandwidth capacities of peer links. Specifically, we formulate the problem of fast replenishment via linear programing framework. Our simulation results indicate that the proposed fast replenishment significantly outperforms the current approach under many typical network scenarios.
