Speaker

Yi Liu

Abstract

Cloud computing has revolutionized how we manage and process data, yet traditional database architectures struggle to fully leverage the benefits of cloud infrastructure. This talk explores the design and optimization of disaggregated database systems, where computational and storage resources are separated into distinct pools and interconnected via high-speed networks to maximize resource utilization and scalability. This talk will discuss two key approaches for building high-performance disaggregated databases: (i) Tailored RDMA-based disaggregated key-value store. (ii) Decoupled data service units to achieve flexible data placement in data centers while maintaining efficient memory usage for indexing. Last, I will discuss future directions to further contribute to high-performance and modern distributed storage and disaggregation in the massive-scale data processing era.

Bio

Yi Liu is a final-year PhD student in the Department of Computer Science and Engineering at the University of California Santa Cruz, working with Prof. Chen Qian. He obtained his Bachelor's degree from the University of Science and Technology of China in 2020. His research mainly focuses on computer networking, storage systems, and databases, especially for building disaggregated database systems in data centers. His work has been published in top-tier conferences, such as VLDB, SIGMETRICS, ICNP, IEEE CLOUD, and ACM/IEEE SEC. His work SMASH has been selected and republished as a research highlight in the Communications of ACM.