I’m currently a third-year graduate student at the ShenZhen Institute of Advanced Technology (SIAT), affiliated with the University of Chinese Academy of Sciences (UCAS). I am advised by Prof. Meng Jintaoand Prof. Wang Yang in SIAT. My primary research collaborations are with Prof. Dr.-Ing. André Brinkmann at Johannes Gutenberg - Universität Mainz. Together, we have been working on a combination of computer systems and databases, especially for key-value store optimization.

I am currently looking forward to a PhD position.

Research Interests

Data-driven Database Tuning

Utilizing stochastic planning, machine learning (ML), or LLM to auto-tune data structures (e.g., B⁺-trees, LSM-trees) to ensure optimal performance across diverse read/write workloads.

NoSQL Data Management Systems

Optimizing the NoSQL databases (DBs), including Key-Value (KV) stores, Graph DBs, and Document DBs, while developing new structures (e.g., new LSM-like structures) to improve read/write performance and scalability. LuMDB.

Modern Hardware Acceleration

Leveraging (not based) cutting-edge hardwares, e.g., NVMe SSDs, persistent memory (PM), and Zone-namespace (ZNS) SSDs, to enhance performance in:

• File System Expansion: Expanding the capabilities of file systems (e.g., EXT4) to support better high-performance apps (e.g., KV stores, RDBs).
• Native Storage Optimization: Application-specific (e.g., KV store) optimizations on native storage. REXIO.

Selected Publications

2024

1. “REXIO: Indexing for Low Write Amplification by Reducing Extra I/Os in Key-Value Store under Mixed Read/Write Workloads”, Zizhao Wang,Jintao Meng^*,Nan Han,Zhelang Deng,Yizhuo Ma,Xiaowen Huang, International Web Information Systems Engineering conference, main Track (WISE main), 2024. PDF Code
2. “Low Unleashes More: Separating low-Access Data for Optimal Read/Write in LSM-based Key-Value Stores”, Zizhao Wang,Yunjue Gu,Wenhan Feng,Jintao Meng,Yang Wang,André Brinkmann, 2024. PDF Code