On Scheduling and Redundancy for P2P Backup
An online backup system should be quick and reliable in both saving and restoring users' data. To do so in a peer-to-peer implementation, data transfer scheduling and the amount of redundancy must be chosen wisely. The authors formalize the problem of exchanging multiple pieces of data with intermittently available peers, and they show that random scheduling completes transfers nearly optimally in terms of duration as long as the system is sufficiently large. Moreover, they propose an adaptive redundancy scheme that improves performance and decreases resource usage while keeping the risks of data loss low.