About Me
I am a tenure-track Associate Professor at Nanjing University. I received my Ph.D. degree from Tsinghua University in 2023 (Advisor: Prof. Jiwu Shu), and Bachelor's degree from Huazhong University of Science and Technology in 2018. From 2023 to 2025, I worked as a Postdoctoral Researcher at Tsinghua University.
Prospective students: We are seeking highly motivated PhD students, Master’s students, and interns. If you're passionate about system research and interested in working with me, please feel free to reach out via email.
Research
My research spans computer systems, from operating systems to distributed systems. My current research seeks to answer the following three questions:
- How to augment computer systems with rapidly advancing hardware technologies
- How to accelerate the evolution of computer systems through large-scale simulation and agentic automation
- How to build efficient and dependable system foundations for long-horizon agents
Some of my past research projects include:
-
Network-Accelerated Distributed Systems
- Leveraging NIC hardware ACKs for fast commit (NSDI'25)
- Network primitives for replication (OSDI'23)
- Distributed locks (ISCA'26, FAST'25, FAST'23)
- Transactions (USENIX ATC'22, VLDB'22)
- In-network cache coherence (FAST'21)
-
Emerging Memory Systems
- Index on disaggregated memory (EuroSys'25, SIGMOD'22)
- Persistent memory systems (OSDI'21, USENIX ATC'22/ATC'23)
- SSD-extended remote memory (FAST'24)
- Protective remote memory (FAST'23)
Services
Program Committee:
- USENIX Conference on File and Storage Technologies (FAST 2027)
- SIGOPS Annual Technical Conference (ATC 2026)
- European Conference on Computer Systems (EuroSys (fall) 2026)
Journal Reviewing:
- TOS, TACO, CSUR
Publications
2027
- [NSDI'27]ns-GPU: Towards Fast Simulation for Large-Scale Networks with GPUs
The 24th USENIX Symposium on Networked Systems Design and Implementation, 2027 - [HPCA'27]Hardware/Software Co-Design for Efficient RPCs over Non-Cache-Coherent Shared CXL Memory
The 33rd IEEE International Symposium on High-Performance Computer Architecture, 2027 - [PPoPP'27]DualDM: Accelerating Disaggregated Memory Applications by Prioritizing Contended Critical Sections
The 32nd ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2027 - [PPoPP'27]CacheChord: Unleashing the Internal Parallelism for LLM serving on Flash Storage
The 32nd ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2027 - [EuroSys'27]Beyond Overlap: Hierarchical Graph Abstraction for Out-of-Core Vector Search
Proceedings of European Conference on Computer Systems, 2027 - [EuroSys'27]NetMoE: Orchestrating Fine-Grained Overlap for Efficient Distributed MoE Inference
Proceedings of European Conference on Computer Systems, 2027 - [TOS'27]Tape in the Cloud: A Cost-Efficient Archive Storage System
ACM Transactions on Storage, 2027 (Fast-tracked)
2026
- [FAST'26]Cost-efficient Archive Cloud Storage with Tape: Design and Deployment
The 24th USENIX Conference on File and Storage Technologies, 2026 (Best Paper Award Candidate) - [ISCA'26]Breaking Barriers in Atomic Scaling: A Hardware-Software-Collaborated Framework to Deconstruct RDMA Atomic
Proceedings of International Symposium on Computer Architecture, 2026 - [MICRO'26]Hera: Eliminating the I/O Tax of RAID on High-Capacity SSDs
The 59th IEEE/ACM International Symposium on Microarchitecture, 2026 - [ASE'26]AlphaCu: A Transformation-Driven Synthesis Framework for LLM-based GPU Kernel Generation
The 41st IEEE/ACM International Conference on Automated Software Engineering, 2026 - [SIGCOMM'26]Pegasus: A Data Center Network for Bare-Metal AI Cloud
ACM SIGCOMM 2026 Conference, 2026 - [ICDE'26]Switch∆: Asynchronous Metadata Updating for Distributed Storage with In-Network Data Visibility
The 42nd IEEE International Conference on Data Engineering, 2026 - [HPDC'26]BCCE: Block-Centric GPU Co-Design for Real-Time Range-Top-K Query at Scale
ACM International Symposium on High-Performance Parallel and Distributed Computing, 2026 - [HPDC'26]STAR: Decode-Phase Rescheduling for LLM Inference
The 35th International Symposium on High-Performance Parallel and Distributed Computing, 2026
2025
- [NSDI'25]Achieving Wire-Latency Storage Systems by Exploiting Hardware ACKs
The 22nd USENIX Symposium on Networked Systems Design and Implementation, 2025 - [CACM]Designing an Efficient Tree Index on Disaggregated Memory [Code]
Communications of the ACM, 2025 (Research Highlight) - [SoCC'25]Accelerating Distributed Filesystem Metadata Service via Decoupling Directory Semantics from Metadata Indexing
ACM Symposium on Cloud Computing, 2025 - [USENIX ATC'25]Weaver: Efficient Multi-LLM Serving with Attention Offloading
USENIX Annual Technical Conference, 2025 - [FAST'25]ShiftLock: Mitigate One-sided RDMA Lock Contention via Handover [Code]
The 23rd USENIX Conference on File and Storage Technologies, 2025 (Distinguished Artifact Award) - [EuroSys'25]Deft: A Scalable Tree Index for Disaggregated Memory [Code]
The 20th European Conference on Computer Systems, 2025 - [TOS]Efficiently Enlarging RDMA-Attached Memory with SSD [Code]
ACM Transactions on Storage, 2025 (Fast-tracked) - [TC]StageWise: Accelerating Persistent Key-Value Stores by Thread Model Redesigning
IEEE Transactions on Computers, 2025 - [HotStorage'25]CableCache: In-Network Request Deduplication for Key-Value Stores
The 17th ACM Workshop on Hot Topics in Storage and File Systems, 2025
2024
- [FAST'24]TeRM: Extending RDMA-Attached Memory with SSD [Code]
The 22nd USENIX Conference on File and Storage Technologies, 2024
Invited for fast-tracked publication in ACM Transactions on Storage - [TOS]Perseid: A Secondary Indexing Mechanism for LSM-based Storage Systems [Code]
ACM Transactions on Storage, 2024 (Fast-tracked) - Challenges and Technical Development of Large Model Training Storage Systems
Journal of Computer Research and Development (in Chinese), 2024
2023
- [OSDI'23]Replicating Persistent Memory Key-Value Stores with Efficient RDMA Abstraction [Code]
The 17th USENIX Symposium on Operating Systems Design and Implementation, 2023 - [SIGMOD Record]Building Write-Optimized Tree Indexes on Disaggregated Memory [Code]
SIGMOD Record, March 2023 (Vol. 52, No. 1), 2023 (Invited) - [USENIX ATC'23]Revisiting Secondary Indexing in LSM-based Storage Systems with Persistent Memory [Code]
USENIX Annual Technical Conference, 2023
Invited for fast-tracked publication in ACM Transactions on Storage - [FAST'23]Patronus: High-Performance and Protective Remote Memory [Code]
The 21st USENIX Conference on File and Storage Technologies, 2023 - [FAST'23]Citron: Distributed Range Lock Management with One-sided RDMA
The 21st USENIX Conference on File and Storage Technologies, 2023 - [VLDB'23]PetPS: Supporting Huge Embedding Models with Persistent Memory [Code]
The 49th International Conference on Very Large Data Bases, 2023 - Survey on In-Network Storage Systems
Journal of Computer Research and Delevopment (in Chinese), 2023 - Progress on Storage Systems for Disaggregated Data Centers
SCIENTIA SINICA Informationis (in Chinese), 2023
2022
- [SIGMOD'22]Sherman: A Write-Optimized Distributed B+Tree Index on Disaggregated Memory [Code]
ACM SIGMOD International Conference on Management of Data, 2022
2023 ACM SIGMOD Research Highlight Award, CACM Research Highlight - [TOS]Nap: Persistent Memory Indexes for NUMA Architectures [Code]
ACM Transactions on Storage, 2022 (Fast tracked) - [USENIX ATC'22]AlNiCo: SmartNIC-accelerated Contention-aware Request Scheduling for Transaction Processing
USENIX Annual Technical Conference, 2022 - [USENIX ATC'22]Pacman: An Efficient Compaction Approach for Log-Structured Key-Value Store on Persistent Memory [Code]
USENIX Annual Technical Conference, 2022 - [EuroSys'22]Fleche: An Efficient GPU Embedding Cache for Personalized Recommendations
The 17th European Conference on Computer Systems, 2022 - [VLDB'22]SwitchTx: Scalable In-Network Coordination for Distributed Transaction Processing
The 48th International Conference on Very Large Data Bases, 2022
2021
- [OSDI'21]Nap: A Black-Box Approach to NUMA-Aware Persistent Memory Indexes [Code]
The 15th USENIX Symposium on Operating Systems Design and Implementation, 2021
Invited for fast-tracked publication in ACM Transactions on Storage - [FAST'21]Concordia: Distributed Shared Memory with In-Network Cache Coherence
USENIX Conference on File and Storage Technologies, 2021 - [TOS]Octopus+: an RDMA-enabled Distributed Persistent Memory File System
ACM Transactions on Storage, 2021 - [ICDE'21]Aria: Tolerating Skewed Workloads in Secure In-memory Key-value Stores
37th IEEE International Conference on Data Engineering, 2021 - A Multicore-Friendly Persistent Memory Key-Value Store
Journal of Computer Research and Delevopment (in Chinese), 2021
2020
- [DAC'20]Improving the Concurrency Performance of Persistent Memory Transactions on Multicores
Design Automation Conference, 2020 - [ASPLOS'20]FlatStore: an Efficient Log-Structured Key-Value Storage Engine for Persistent Memory
Proceedings of the 25th International Conference on Architectural Support for Programming Languages and Operating Systems, 2020 - [VLDB'20]μTree: a Persistent B+-Tree with Low Tail Latency
46th International Conference on Very Large Data Bases, 2020 - [TOS]TH-DPMS: Design and Implementation of an RDMA-enabled Distributed Persistent Memory Storage System
ACM Transactions on Storage, 2020
Selected Awards
- Best Paper Award Candidate, USENIX FAST 2026
- Distinguished Artifact Award, USENIX FAST 2025
- CACM Research Highlight, 2025
- ACM SIGMOD Research Highlight Award, 2023
- Microsoft Research PhD Fellowship, 2021
- Huawei OlympusMons Award, 2020