Workshop:Fifth International Parallel Data Systems Workshop
Authors: Huan Ke and Haryadi S. Gunawi (University of Chicago) and Dominic Manno, David Bonnie, and Bradley W. Settlemyer (Los Alamos National Laboratory)
Abstract: In this paper, we propose a flexible and practical data protection scheme, fractional-overlap declustered parity( FODP), to explore the trade-offs between fault tolerance and rebuild performance. Our experiments show that FODP is able to bring forth up to 99% less probability of data loss in the presence of various failure regimes. Furthermore, by adding one additional spare drive capacity within each server, FODP yields up to 99% reduction in granularity of data loss.