Data Dependent Sparing to Manage Better-Than-Bad Blocks




Abstract: We forecast that bad block management will remain critical for future systems built with advanced and emerging memory technologies. We argue that the conventional block retirement and sparing approach—a block is retired as soon as it shows faulty behavior—is overly conservative. We observe that it is highly unlikely that all faulty bits in a block manifest errors. Consequently, we propose data dependent sparing, a relaxed block retirement and sparing approach that recycles faulty storage blocks.

Bio: Rakan Maddah is a Ph.D student in the Computer Science Department at the University of Pittsburgh. He joined Pitt in Fall 2010 after earning an M.Sc. in Computer Science from the Lebanese American University. His research interests are in computer architecture and fault tolerance. He is currently working on efficient techniques to alleviate the endurance problem of Phase Change Memory and extend its lifetime.