Presenter: Rashmi Vinaya, Accociate Professor in Computer Science, Carnegie Mellon University

Description: Cluster storage systems, the foundation of all data systems, store large amounts of data (in the range of Petabytes to Exabytes) spread across millions of hard disk drives. Due to their scale, they are costly and consume significant resources. It is therefore crucial to make them resource-efficient. At the same time, disk failures are common, and data must be protected through redundancy, which is resource-intensive.

In this talk, I will present techniques for orchestrating redundancy in cluster storage systems, called DARE (Disk-Adaptive REdundancy), that significantly reduce resource overhead while also improving data protection guarantees. Specifically, I will (1) present an analysis of production failure traces (including those from Google) showing that storage clusters consist of disks with widely varying failure rates, (2) show that substantial space-savings can be realized by adapting redundancy schemes to observed failure rates, (3) present challenges and solutions for achieving such an adaptation in a practical way that can be deployed in production systems, (4) introduce a new class of coding-theoretic tool for adding redundancy called Convertible codes, which enable efficient redundancy adaptation with theoretical guarantees.

Bio:  Rashmi Vinayak is an associate professor in the Computer Science department at Carnegie Mellon University. Rashmi received her Ph.D. from UC Berkeley in 2016, and was a postdoctoral scholar at UC Berkeley from 2016-17. Rashmi is a recipient of Sloan Research Fellowship 2023, Meta Research Award 2022, VMware Systems Research Award 2021, NSF CAREER Award 2020-25, Tata Institute of Fundamental Research Memorial Lecture Award 2020, Facebook Distributed Systems Research Award 2019, Google Faculty Research Award 2018, and Facebook Communications and Networking Research Award 2017. Her PhD thesis was awarded the UC Berkeley Eli Jury Dissertation Award 2016, and her work has received USENIX NSDI 2021 Community (Best Paper) Award, and IEEE Data Storage Best Paper and Best Student Paper Awards for the years 2011/2012. During her Ph.D. studies, Rashmi was a recipient of Facebook Fellowship 2012-13, the Microsoft Research PhD Fellowship 2013-15, and the Google Anita Borg Memorial scholarship 2015-16. Her research interests broadly lie in computer/networked systems and information/coding theory, and the wide spectrum of intersection between the two areas.

Hosted by: Professor Steve McGuire

Zoom link:  https://ucsc.zoom.us/j/94057173213?pwd=SFFnQWV3ai84UFJXclpmNEJvWkY1UT09