Xinjing Zhou(周信静)

I am a Postdoc in MIT’s Data Systems Group, working with Samuel Madden on agentic data systems. I completed my PhD in the same group, advised by Mike Stonebraker. My research focuses on building data systems for AI agents: long-running, stateful, and often speculative applications that interact with databases, file systems, vector stores, and external tools. More broadly, my work spans database systems, operating systems, and systems for AI, drawing on my PhD work on practical DB–OS co-design, kernel bypass, memory management, networking, and high-performance database engines.

Email  /  Google Scholar  /  Twitter  /  Linkedin  /  Github

profile photo
Publications
Tux: Efficient Drop-in Networking for Database Systems
Xinjing Zhou, Viktor Leis, Xiangyao Yu, Michael Stonebraker
To appear in VLDB 2026

Consistency and Correctness in Data-Oriented Workflow Systems
Michael Stonebraker, Xinjing Zhou, Peter Kraft, Qian Li
Proceedings of The Conference on Innovative Data Systems Research (CIDR), January 2026

Tiered-Indexing: Optimizing Access Methods for Skew
Xinjing Zhou , Xiangpeng Hao, Xiangyao Yu, Michael Stonebraker
The VLDB Journal 34, 45 (2025)
code

Practical DB-OS Co-Design with Privileged Kernel Bypass
Xinjing Zhou, Viktor Leis, Jinming Hu, Xiangyao Yu, Michael Stonebraker
Proceedings of the ACM on Management of Data, 2025, Volume 3, Issue 1, Article 64
slides/ code

OLTP Through the Looking Glass 16 Years Later: Communication is the New Bottleneck
Xinjing Zhou, Viktor Leis, Xiangyao Yu, Michael Stonebraker
Proceedings of The Conference on Innovative Data Systems Research (CIDR), January 2025
slides

Towards Buffer Management with Tiered Main Memory
Xiangpeng Hao, Xinjing Zhou, Xiangyao Yu, Michael Stonebraker
Proceedings of the ACM on Management of Data, 2024, Volume 2, Issue 1, pp 1–26

Epoxy: ACID Transactions across Diverse Data Stores
Peter Kraft, Qian Li, Xinjing Zhou, Peter Bailis, Michael Stonebraker, Matei Zaharia, Xiangyao Yu
Proceedings of the VLDB Endowment, 2023, Volume 16, Issue 11, pp 2742–2754

Two is Better Than One: The Case for 2-Tree for Skewed Data Sets
Xinjing Zhou, Xiangyao Yu, Goetz Graefe, Michael Stonebraker
Proceedings of The Conference on Innovative Data Systems Research (CIDR), January 2023.
video / code / slides

Lotus: Scalable Multi-Partition Transactions on Single-Threaded Partitioned Databases
Xinjing Zhou, Xiangyao Yu, Goetz Graefe, Michael Stonebraker
Proceedings of the VLDB Endowment, 2022
code/ slides

Spitfire: A Three-Tier Buffer Manager for Volatile and Non-Volatile Memory
Xinjing Zhou, Joy Arulraj, Andy Pavlo, David Cohen
Proceedings of the SIGMOD, 2021
video / code

DPTree: Differential Indexing for Persistent Memory
Xinjing Zhou, Lidan Shou, Ke Chen, Wei Hu, Gang Chen
Proceedings of the VLDB Endowment, 2020
video / code

Talks

Practical DB-OS Co-design with Privileged Kernel Bypass, Salesforce Database Reading Group, June 2025

Practical DB-OS Co-design with Privileged Kernel Bypass, Columbia DB Seminar, March 2025

OLTP Through the Looking Glass 16 Years Later, Microsoft Gray Systems Lab, March 2025

Services

  • External reviewer for SIGMOD 2021, VLDB 2021
  • External reviewer for VLDBJ 2022, VLDBJ 2024.


  • Template from jonbarron.