Hao Chen (陈浩)

I obtained my Ph.D. degree in Computer Science from University of Science and Technology of China (USTC) in 2021, guided by Professors Yinlong Xu and Cheng Li. During my Ph.D. studies, I interned at Qatar Computing Research Institute (QCRI) as a research assistant from 2019 to 2021, mentored by Xiaosong Ma.

Currently, I’m part of the PolarDB MySQL Department at Alibaba Cloud, focusing on the development and research of next-generation cloud-native databases. My research interests encompass cloud-native databases, RDMA (Remote Direct Memory Access), and CXL, with a particular focus on their evolving advancements and practical applications. I am also expanding my research interests toward AI infrastructure, with a focus on scalable model serving, KV cache management, and efficient resource utilization. I am keenly interested in exploring innovative partnerships and research opportunities in this field.

Research Intern Openings: We are looking for research interns interested in AI-native data systems and AI infrastructure. Potential topics include vector databases, agent memory systems, KV cache management, RDMA/CXL-based disaggregated memory systems, and efficient LLM inference serving. If chasing down bottlenecks in large-scale cloud and AI infrastructure sounds like fun, we would be happy to hear from you.

Feel free to reach out via email: cighao@gmail.com.

Awards

  • SIGMOD 2025 Industry Track Best Paper Award (Corresponding Author)
  • SIGMOD 2024 Industry Track Best Paper Award (Corresponding Author)
  • ICDE 2024 Best Industry and Application Paper Award (Corresponding Author)

Experience

Alibaba Cloud, Jul 2021 - Present

Research and development of next-generation cloud-native databases and AI infrastructure.

  • KVCache for LLM Serving [PVLDB'26]
  • Cloud-Native Database Architecture: PolarDB Multi-Primary [SIGMOD'24 Best Paper], PolarDB Serverless [ICDE'24 Best Paper]
  • Distributed Database Systems: PolarDB Strongly Consistent Reads [PVLDB'23], PolarDB Limitless [PVLDB'25], Geo-Replication [PVLDB'24]
  • Memory-Centric Database Systems: Persistent Memory [ASPLOS'23], CXL Memory Pool [SIGMOD'25 Best Paper], CXL-based Locking [TC'26], AP Memory Pool [SIGMOD'26]
Qatar Computing Research Institute, Apr 2019 - Jan 2021

Ph.D. Research Intern.

  • SpanDB: Hybrid-Storage KV Store [FAST'21]
  • QarSUMO: Parallel Traffic Simulator [SIGSPATIAL'20]

Publications

2027 FAST

CakeMask: Enabling Fresh-Data HTAP Databases with High Performance and Low Memory Overhead

Baoyue Yan, Jianying Wang, Hao Chen, Xinjun Yang, Feifei Li

25th USENIX Conference on File and Storage Technologies (FAST). 2027. To appear.

2026 PVLDB

PolarKV: Tier Locally, Serve Globally—A KV Cache over Cloud Memory and Storage

Yingqiang Zhang, Xinjun Yang, Hao Chen, Feifei Li, Shuwen Wang, Chuan Sun, Sai Wu, Ninglong Weng. ( corresponding author)

Proceedings of the VLDB Endowment (PVLDB). 2026. [PDF]

Conferences & Talks

Places, people, and ideas along my research journey.

  1. [Sponsor Talk] Where Data Gravity Meets Intelligence: Building AI-Native Databases for Agentic EraVLDB 2026Boston
  2. PolarKV: Tier Locally, Serve Globally—A KV Cache over Cloud Memory and StorageVLDB 2026Boston
  3. Towards a Shared-Storage-Based Serverless Database Achieving Seamless Scale-Up and Read Scale-OutICDE 2024Utrecht
  4. HCFTL: A Locality-Aware Page-Level Flash Translation LayerDATE 2019Florence