Please note: This seminar will be given online.
Sudarsun
Kannan,
Department
of
Computer
Science
Rutgers
University
The last decade has seen a rapid hardware innovation to develop ultra-fast and heterogeneous storage and memory technologies for accelerating data-intensive applications. Unfortunately, current monolithic system software stacks with coarse-grained synchronization, high data movement costs, and inflexible abstractions continue to be Achille’s heel, thereby failing to exploit hardware innovations.
In the first part of the talk, I will discuss our efforts to design a cross-layered and scalable storage system disaggregated across the host and the storage device layers for exploiting compute and device parallelism. In the second part of the talk, I will discuss the software overheads and data placement challenges in heterogeneous memory systems and our baby steps to address these challenges. Finally, I will conclude the talk by outlining unsolved challenges and prospective future directions.
Bio: Sudarsun is an assistant professor at Rutgers University, where he leads the Rutgers Systems Lab. His research group works at the intersection of hardware and software, building operating systems and system software for next-generation memory and storage technologies. Results from his work have appeared at premier operating systems and architecture venues, including OSDI, SOSP, EuroSys, FAST, ISCA, and others. Sudarsun’s work has also resulted in patents related to nonvolatile memory and resource management. Before joining Rutgers, he was a postdoctoral research associate at Wisconsin-Madison and graduated with an M.S. and Ph.D. from Georgia Tech.
To join this seminar on Zoom, please go to https://uwaterloo.zoom.us/j/99449197983?pwd=NDFCbHV3dWZ3T1U5eXZXTTdhTU9qZz09.