Sky Seminar Series: Rashmi Vinayak (CMU) – Co-designing Systems and Algorithms for Efficient and Performant Data Systems

Speaker: Rashmi Vinayak

Location: Soda 510

Date: March 8, 2024

Time: 12-1pm PST

Title: Co-designing Systems and Algorithms for Efficient and Performant Data Systems


Data systems, encompassing storage, caching, and processing, are the foundation of all data-intensive applications. They handle vast amounts of data and process a massive number of user requests, all while aiming for low latency, high throughput, and resilience. Their scale and constant operation lead to significant resource and energy consumption. In this talk, I will present my group’s work on system design and algorithmic innovations in storage, caching, and ML systems, aimed at enhancing their efficiency, performance, and reliability.

The first part of the talk will focus on cluster storage systems. I will present techniques for dynamically adapting redundancy levels to achieve substantial resource savings while also strengthening data protection. This discussion will include insights from real-world production failure traces (including those from Google), and  a discussion on practical challenges and solutions for achieving adaptation in production systems. The second part of the talk highlights our recent research on caching and ML systems. Several of the works that I will present have been adopted by industry, both in large-scale production systems and in popular open-source libraries.

Bio: Rashmi Vinayak is an associate professor in the Computer Science department at Carnegie Mellon University. Her research interests broadly lie in computer/networked systems and information/coding theory, and the wide spectrum of intersection between the two areas. 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 several awards including Sloan Research Fellowship, IEEE ITSoc Goldsmith Lecturer award, VMware Systems Research Award, NSF CAREER Award, TIFR Memorial Lecture Award, multiple research awards from Meta and Google, and UC Berkeley Eli Jury Dissertation Award. Her work has been adopted by industry, including at Google, Meta, Twitter, VMware, and several open source libraries, received USENIX NSDI Community (Best Paper) Award, and IEEE Data Storage Best Paper and Best Student Paper Awards, and featured on popular media platforms including HackerNews. During her Ph.D. studies, Rashmi was a recipient of the Facebook Fellowship, the Microsoft Research PhD Fellowship, and the Google Anita Borg Memorial Scholarship.