CANCELLDED ICSA Colloquium - 23/08/19

Title

Rethinking System Support for Persistent Memory

Abstract

For decades our systems have been designed to leverage the two-level storage, where data is persisted in non-volatile storage, but the processor manipulates data in volatile main memory. Recent non-volatile memory (NVM) technologies can manipulate persistent data directly in memory. In this talk,  I will present my vision to redefine the hardware and system stack to unify memory and storage with persistent memory. The advancement of NVM memory technology has spurred the development of crash-consistent applications for persistent memory. On top of that, persistent memory needs system support for encryption, integrity protection, compression, deduplication, etc., necessary to provide security, endurance, and lifetime guarantees. This talk focuses on the challenges in providing correct and efficient system support for persistent memory and fundamentally rethinks how we design our systems with unified memory and storage.

Biography

Samira Khan is an Assistant Professor at the University of Virginia (UVa). Her group ShiftLab focuses on building system stack for emerging technologies. Prior to joining UVa, she was a Post Doctoral Researcher at Carnegie Mellon University, funded by Intel Labs. She is the recipient of NSF CRII Award and Rising Stars in EECS Award. She received her PhD from the University of Texas at San Antonio. During her graduate studies, she worked at Intel, AMD, and EPFL.