Peixiong He Defends and Joins Tuskegee University
In summer 2026, Peixiong He successfully defended a doctoral dissertation titled An Energy-Aware SQL Database System. Currently, Dr. He is an Assistant Professor of Computer Science at Tuskegee University, AL.
Dr. He received the Ph.D. degree in Computer Science and Software Engineering from Auburn University in August 2026, completing a nine-year academic journey at Auburn.
Peixiong He received the B.S. and M.S. degrees in Computer Science from Auburn University in 2021 and 2025. During graduate study, Dr. He served as a graduate teaching assistant and a graduate research assistant in the Department of Computer Science and Software Engineering, and received the Charles E. Gavin Fellowship Award in 2021. Dr. He's research interests include Data Mining, Database Systems, Data Storage Systems, Machine Learning, Energy-Efficient Systems, Dynamic Resource Management, and Recommendation Algorithms.
In the past few years, Dr. He has been working on several projects related to energy-efficient database systems. In the dissertation study, Dr. He developed GreenDB_DM, an energy-aware SQL database system that uses workload-aware data mining to decide which tables to keep on powered-on nodes. The first component, GARMT, groups consecutive SQL queries and applies a modified FP-Growth algorithm to discover tables that are frequently accessed together, reducing pattern-mining time by up to 40% while preserving rule quality. GreenDB_DM then places complete groups of co-accessed tables on the hot node, so that entire queries can be served without waking up cold nodes. The evaluation results show that GreenDB_DM improves the full-query hit rate by 5 to 15 percentage points and reduces total cluster energy by an additional 2.4% to 4.7% beyond GreenDB. Dr. He also modeled recommender systems as disease-spread processes on social networks, showing how recommendation strategies produce different propagation patterns over time.
Click here to read the GARMT paper published in Computers, and here to read the paper on modeling recommender systems published in Information.