Title: Enhance Database Query Performance through Software Optimization and Hardware Adaptation

Date: 2/19 (Monday)

Time: 11 AM - 12:30 PM

Location: Klaus 3100

Virtual: Team Link

Join conversation

teams.microsoft.com

Jiashen Cao

Ph.D. Computer Science

School of Computer Science

Georgia Institute of Technology

Committee:

Dr. Joy Arulraj (Co-Advisor) - School of Computer Science, Georgia Institute of Technology

Dr. Hyesoon Kim (Co-Advisor) - School of Computer Science, Georgia Institute of Technology

Dr. Divya Mahajan - School of Electrical and Computer Engineering and Computer Science, Georgia Institute of Technology

Dr. Kexin Rong - School of Computer Science, Georgia Institute of Technology

Abstract:

Database systems have been incorporating machine learning algorithms to provide rich information. Complex machine learning algorithms become the new performance bottleneck for database systems, with video database systems particularly suffering from this performance degradation. In this talk, I will discuss my endeavors to enhance query performance through software optimizations and hardware adaptations.

I first present FiGO, an approach to utilize the just-enough-accurate algorithm to process data. Following that, I will introduce Hydro, which employs the concept of adaptive query processing. This strategy allows for executing queries with optimal planning and resource allocation. I will also cover my research efforts in analyzing and optimizing GPU database systems. Lastly, I will present my proposed research idea of leveraging existing deep learning compilers to construct an optimal query plan for machine learning queries.