Zhaoheng “Billy” Li

I am a third-year CS PhD student at the University of Illinois at Urbana-Champaign. I am a member of CreateLab advised by Prof. Yongjoo Park.

My research interests lie within Database Systems on the topics of Workload optimization and Latency reduction for ad-hoc analytical queries. My recent work focuses on capturing and utilizing both client and server-side metrics from previous/recurring workloads to speedup future workloads through a combination of caching and pre-computation. I frequently employ techniques from graph mining and combinatorial optimization in my work.

My current work:

• Fast dataframe operations via optimally scheduling CPU and GPU usage
• Kishu: fast incremental checkpointing and versioning for machine learning and EDA notebook workloads

My previous research projects:

• ElasticNotebook: Enabling Live Migration for Computational Notebooks (VLDB 2024, with Pranav Gor, Rahul Prabhu, Hui Yu, Yuzhou Mao, and Prof. Yongjoo Park)
• Transactional Python for Durable Machine Learning: Vision, Challenges, and Feasibility (SIGMOD DEEM 2023, with Supawit Chockchowaat and Prof. Yongjoo Park
• S/C: Speeding up Data Materialization with Bounded Memory (ICDE 2023, with Xinyu Pi, and Prof. Yongjoo Park)
• REFORM: Fast and Adaptive Solution for Subteam Replacement (IEEE BigData 2022, with Xinyu Pi, Mingyuan Wu, and Prof. Hanghang Tong)
• Deep Steerable Graph Generation with Xinyu Pi, Yuheng Chang, and Prof. Carl Yang

I am a four-time software engineer intern at Google:

• Summer 2023: Google Cloud, Google BigQuery, Group by Struct with Xueping Weng and Joel Wasserman
• Summer 2022: Google Cloud, DAS-Performance, SQL query profiling with BPF with Tengyu Sun and Hao Luo
• Summer 2020: Google Ads, Google Local Services, improved ranking algorithms for Home Services Ads with Dr. Bharath Pattabiraman
• Summer 2019: Google Ads, ContentAds Team, monitoring service for ContentAds requests with Dr. Darek Yung