Colloquium: Fast and Stable Data and Storage Systems in Milli-/Micro-second Era
Abstract: As the need for real-time data surges, storage systems, the home of data, are facing a significant challenge of maintaining performance sustainability and offering low and stable response times. With various demands, intricate structures, and emerging storage hardware, our systems become more vulnerable to tail latencies, which can cause bad user experience and severe revenue loss, highlighting the necessity for new technologies. In this talk, I will introduce my view on how to make layers in our hardware-software co-stack work together to tackle this challenge from multiple angles, and eventually make our data and storage systems “tail-free”.
Bio: Mingzhe Hao is a Ph.D. candidate in the Department of Computer Science at the University of Chicago. His research interests are in combining heuristics methods and machine learning techniques to build solutions for various systems, such as storage systems, operating systems, and cloud/distributed systems. Specifically, he builds fast and stable systems that achieve rapid responses even in the most turmoil scenarios.
He publishes his work in conferences such as ASPLOS, FAST, OSDI, SOSP, and SoCC. Some publications he is involved in are nominated for the best paper awards for their potential impacts. Meanwhile, his expertise has allowed him to serve as a reviewer/external reviewer for conferences and journals including FAST, JPDC, TC, and TOS. Also, as a recognition of his academic merit, he has received multiple honors/awards including Siebel Scholar (class 2020), William Rainey Harper Dissertation Fellowship, Elastos Fellowship, and Facebook Fellowship Finalist.
*To sign up for an appointment to meet with Mingzhe Hao, please pay a visit to the following link: Appointment Sign Up Sheet.