Speaker: Vijay Chidambaram
Location: Soda 510
Date: March 15, 2024
Time: 12-1pm PST
Title:
Building robust storage systems for persistent memory
Abstract:
Persistent memory (PM) is a new class of memory technology that combines the low latency of DRAM with the durability of storage. In this talk, I’ll present my group’s recent work on building robust storage systems for PM. First, I will present Chipmunk, a tool for finding crash-consistency bugs in persistent-memory file systems. Chipmunk works with kernel, user-space, and hybrid PM file systems. I’ll discuss the bugs that have been discovered using Chipmunk, and the lessons learned about developing file systems for PM. Second, I will talk about our work on developing a PM file system in Rust. Using specific properties of an unmodified Rust compiler, we are able to statically verify ordering-related file-system invariants. This allows building a PM file system with significantly stronger compile-time guarantees than existing state-of-the-art.
Bio:
Vijay Chidambaram is an Associate Professor in the Computer Science department at the University of Texas at Austin. He did his post-doc at the VMware Research Group, and got his PhD with Prof. Remzi and Andrea Arpaci-Dusseau at the University of Wisconsin-Madison. His papers have won Best Paper Awards in Eurosys 2023, ATC 2018, FAST 2018, and FAST 2017. He was awarded the NSF CAREER Award in 2018, SIGOPS Dennis M. Ritchie Dissertation Award in 2016, and the Microsoft Research Fellowship in 2014. Techniques from his work have been incorporated into commercial products, and his work has helped make the Linux kernel more reliable.