DAPPER: a database-inspired approach to persistent memory




Abstract: Non-volatile memory (NVM) collapses the boundaries between the in-memory and secondary storage representations of data structures and enables the programmer to process data directly from an imperative runtime. We present the results of the DAPPER project, which takes a database-inspired approach to NVM. It supports recoverable data structures in NVM at the imperative language level, algorithms optimized for the performance characteristics of the new medium and a runtime to support them.

Bio: Stratis D. Viglas is an Associate Professor in the School of Informatics at the University of Edinburgh. He has a PhD in Computer Science from the University of Wisconsin---Madison and BSc and MSc degrees from the Department of Informatics at the University of Athens, Greece. His research interests are in data management over new hardware deployments, with a focus on storage and query processing technologies.