Research Seminar: Making More with Less: Efficient Caching for Key-value and Object Storage Systems

Feng Chen

Assistant Professor of Computer Science, Louisiana State University

September 14, 2018

3:00 pm

Patrick F. Taylor Hall 3107

 

Abstract

In past years, several important storage technologies have emerged and play a vital role in today's cloud services. Two particularly important ones are key-value store and object based
cloud storage. In both systems, caching is a crucial component and essential to system performance and user experience.

In this talk, we will present our recent works on optimizing caching efficiency as well as performance in such scenarios. We will first introduce our research efforts on optimizing memory efficiency and exploiting hardware potentials for caching a huge amount of key-value data in flash. Then we will present a cost-aware caching scheme for object-based cloud storage, aiming for improving performance and lowering cost. Our studies show that by carefully exploiting the unique device and workload properties, we can significantly enhance the efficiency and performance of caching for providing high-speed data services at a low cost.

Bio

Dr. Feng Chen is currently an Assistant Professor in the Division of Computer Science and Engineering at Louisiana State University. Before joining LSU, Dr. Chen was a research scientist with Circuits and Systems Research at Intel Labs. His research interests are mainly in computer systems with a focus on operating systems, memory and storage systems, and data management in cloud and distributed systems. Dr. Chen received his Ph.D. degree in Computer Science and Engineering from the Ohio State University in 2010. He is a recipient of the Best Paper Award in the 25th ACM International Conference on Supercomputing in 2011, the NSF Faculty Early Career Development Award (CAREER) in 2015, and the LSU Alumni Association Rising Faculty Research Award in 2018.