I am Xiaoke Zhu (Hsiaoko Chu in Wade-Giles romanization), currently a research fellow at Shenzhen Institute of Computing Sciences (SICS). In Sep. 2025, I received my Ph.D. degree at Beihang University (BUAA), under the supervision of Prof. Wenfei Fan. Before that, I received master degree at Yunnan University (YNU) in 2020, where I was advised by Prof. Wei Zhou and Prof. Shaowen Yao. My work focuses on Graph Computing, GPU Accelerated Algorithms, High Performance Data Cleaning, and AI4DB. I welcome collaborations and discussions—feel free to reach out via email.

News

• NOV 2024 Our GPU-Accelerated Graph Cleaning with a Single Machine paper was accepted to SIGMOD 2025.
• NOV 2024 Our A Single Machine System for Querying Big Graphs with PRAM paper was accepted to VLDB 2025.
• OCT 2024 Our HyperBlocker: Accelerating Rule-Based Blocking in Entity Resolution Using GPUs paper was accepted to VLDB 2025.
• OCT 2024 Our Deep Learning Service for Efficient Data Distribution Aware Sorting paper was accepted to IEEE BigData 2024.
• Apr 2023 Our MiniGraph: Querying Big Graphs with a Single Machine paper was accepted to VLDB 2023.
• Feb 2022 Our Deep and Collective Entity Resolution in Parallel paper was accepted to ICDE 2022.
• Sep 2021 I joined Shenzhen Institute of Computing Science (SICS) as a research intern.
• Jul 2021 Our DLB: Deep Learning Based Load Balancing paper was accepted to IEEE CLOUD 2021.

Research Interests

My research focuses on graph computing and databases, optimizing runtime efficiency for shared-memory and CPU/GPU architectures, with publications in SIGMOD, VLDB, ICDE, BigData, and CLOUD. A brief summary of my past work can be found below.

Graph Computing Systems

I have worked on building a high-level programming model and runtime system that can execute applications on shared-memory or out-of-memory architectures with CPUs or GPUs. For out-of-core graph analytics (e.g., PageRank, SSSP), I improved I/O efficiency, and for graph mining (e.g., Graph Data Cleaning, Pattern Matching), I optimized GPU performance. Relevant results were published in [VLDB’23, VLDB’25, SIGMOD’25]

Data Cleaning

I have worked on improving the performance of data cleaning systems on modern hardware like GPU or on distributed cluster. I have also compared different parallel runtime systems for data cleaning, and identified their performance bottlenecks. Relevant results were published in [ICDE’22, VLDB’25, SIGMOD’25]

AI4DB

I have leverages machine learning and deep learning model to improve tasks traditionally handled by human database administrators or classical algorithms, enabling more efficient data processing and resource management. Specially I have designed learned models for sorting, load balancing, and scheduling. Relevant results were published in [CLOUD’21, BigData’24]