about research news publications cv
life
Portrait photo Google Scholar

Yunan Zhang - 张宇楠

Ph.D. Student @ Purdue University

Hi! This is Yunan, a Ph.D. student in Computer Science at Purdue University, advised by Professor Jianguo Wang. I came to Purdue University in 2021 as an M.S. student in Computer Science and transferred to the Ph.D. program in 2023. Previously, I received an M.S. in Computer Engineering from the Illinois Institute of Technology and a bachelor’s degree from Shenzhen University.

I have worked as a Research Intern at Microsoft Research in the DiskANN team.

Please feel free to reach out if you're interested in discussing any research ideas, academic collaborations, or industry opportunities!

🔎 Vector Search • 🗂️ Vector Indexing • Vector Databases

Research

My research focuses on building efficient, high-performance vector search indexes and systems across diverse storage devices and real-world workloads. My research covers the following main areas:

  • Vector Search on Databases: Building and optimizing vector search integrated with databases, with a focus on implementation and performance analysis in PostgreSQL and other DBMSs across real-world workloads. ICDE'24, VLDB'24, CIDR'26.
  • Disk-Based Vector Search: Designing and optimizing disk-based vector indexes for commodity storage (HDDs, SSDs, and cloud block storage), including segment-aware optimizations for LSM-style multi-segment systems.
  • Filtered Vector Search: Developing adaptive search strategies for graph-based vector indexes under diverse filtering conditions and query patterns (e.g., single vs. multiple filters and mixed filter types).
  • Related Vector Search Algorithms: Exploring related techniques for vector search, including merge strategies for graph-based indexes and applications in retrieval-augmented generation (RAG) for LLMs. SIGMOD'26

News + Updates

  • Nov. 2025

    The paper "Efficient Vector Index Merging in Vector Databases" has been accepted in SIGMOD 2026.

    publication
  • Oct. 2025

    The paper "Fast Vector Search in PostgreSQL: A Decoupled Approach" has been accepted in CIDR 2026.

    publication
  • June 2025

    Excited to be joining Microsoft Research as a Research Scientist Intern on the Azure DiskANN team. I'll be working with Harsha Vardhan Simhadri on filtered vector search algorithms. Grateful for the opportunity!

    internship
  • June 2024

    The paper "SingleStore-V: An Integrated Vector Database System in SingleStore" has been accepted in VLDB 2024 industrial track.

    publication
  • April 2024

    The paper "Are There Fundamental Limitations in Supporting Vector Data Management in Relational Databases? A Case Study of PostgreSQL" has been accepted in ICDE 2024.

    publication milestone
  • May 2023

    Started to work with Professor Jianguo Wang in vector search. Big thanks to this opportunity!

    milestone

Publications

See all research
  • VLDB 2026 paper teaser
    DiskV: Disk-based Vector Search Optimized for Low-Cost Storage
    Yunan Zhang, Yinghao He, Jiayi Liu, Jianguo Wang
    Submitted to VLDB 2026

    A disk-based vector index designed to achieve high search performance on commodity storage including HDDs and cloud block storages.

  • SIGMOD 2026 paper teaser
    Efficient Vector Index Merging in Vector Databases
    Chenzhe Jin, Yunan Zhang, Jiayi Liu, Jianguo Wang
    In SIGMOD 2026

    A methods for efficient graph vector index merging.

  • CIDR 2026 paper teaser
    Fast Vector Search in PostgreSQL: A Decoupled Approach
    Jiayi Liu, Yunan Zhang, Chenzhe Jin, Aditya Gupta, Shige Liu, Jianguo Wang
    In CIDR 2026 • Paper

    A decoupled approach for efficient vector search in PostgreSQL with comparable performance to specialized vector databases.

  • VLDB 2024 paper teaser
    SingleStore-V: An Integrated Vector Database System in SingleStore
    Cheng Chen, Chenzhe Jin, Yunan Zhang, Sasha Podolsky, Chun Wu, Szu-Po Wang, Eric Hanson, Zhou Sun, Robert Walzer, Jianguo Wang
    In VLDB 2024 • Paper

    Implemented SingleStore-V, a full-fledged, general-purpose vector database integrated into SingleStore.

  • ICDE 2024 paper teaser
    Are There Fundamental Limitations in Supporting Vector Data Management in Relational Databases? A Case Study of PostgreSQL
    Yunan Zhang, Shige Liu, Jianguo Wang
    In ICDE 2024 • Paper

    A case study of limits and practical solutions for implementing high-performance vector search in relational databases (PostgreSQL).

Teaching

CS 44000 Large Scale Data Analytics: Fall 2026, Fall 2025, Spring 2025, Fall 2024

CS 24000 Programming In C: Spring 2024

CS 25000 Computer Architecture: Fall 2023

Professional Service

Journal Reviewer: TKDE 2025

Visitor Map

Recent Visitors

    zhan4404 [at] purdue [dot] edu

    Press Esc or click anywhere to close