Associate Professor (Tenure Track)
Institute of Parallel and Distributed Systems
School of Computer Science
Shanghai Jiao Tong University
Software Building, 800 Dongchuan Rd., Shanghai, China
Zip/Postal Code: 200240
Email: wxdwfc at sjtu dot edu dot cn

I am a tenure-track Associate Professor at Shanghai Jiao Tong University and a member of the Institute of Parallel and Distributed Systems (IPADS). I received my Ph.D. from Shanghai Jiao Tong University in 2021, advised by Professors Binyu Zang, Haibo Chen, and Rong Chen.

I am looking for self-motivated students interested in building practical, high-performance systems, especially infrastructure for AI. If you are interested in joining my group or pursuing a research internship, please contact me at wxdwfc@sjtu.edu.cn.

Google Scholar

I build system infrastructure that improves the performance, scalability, and resilience of modern applications. My current research centers on scalable AI infrastructure, building on my work in serverless computing, distributed storage, and datacenter networking.

• Scalable AI Infrastructure
  Develop efficient and reliable infrastructure for AI workloads, from LLM request scheduling and elastic serving to GPU checkpointing and cloud storage.
  • LLM request scheduling: LMetric explores simple yet effective methods for scheduling LLM requests at scale. OSDI'26 | Code
  • Elastic LLM serving: BlitzScale enables fast, live model autoscaling; KunServe addresses memory overloading through parameter-centric memory management. OSDI'25 | EuroSys'26
  • GPU checkpoint and restore: PhoenixOS is the first system to support concurrent OS-level GPU checkpointing and restoration using validated speculation for fault tolerance and fast startup. SOSP'25 | Code
  • Data access for AI: AITurbo improves cloud storage I/O for AI jobs by coordinating storage and compute resources; our KVCache study characterizes access patterns to improve cache management. FAST'26 | ATC'25

• Serverless and Disaggregated Systems
  Rethink process creation, state transfer, and remote memory access to make cloud applications more elastic and efficient.
  • MITOSIS: Fast RDMA-codesigned remote fork for serverless computing without provisioned concurrency. OSDI'23
  • DMERGE: Serialization/deserialization-free state transfer in serverless workflows. EuroSys'24 — Best Paper Award
  • ODRP: On-demand remote paging with programmable RDMA. NSDI'25

• RDMA-accelerated Storage and Transactions
  Co-design distributed systems with modern datacenter hardware to reduce the cost of data access, coordination, and transaction processing.
  • DrTM: Fast distributed transactions using RDMA and HTM, with replication-driven live reconfiguration and hybrid RDMA primitives. SOSP'15 | EuroSys'16 | ATC'17 | OSDI'18
  • XStore and DST: Learned caching for RDMA-based ordered key-value stores, and decentralized scalar timestamps for efficient MVCC. OSDI'20 | NSDI'21
  • Remote persistent memory: Characterizing and optimizing RDMA access to persistent memory. ATC'21
  • RDMA control and SmartNICs: KRCORE provides a microsecond-scale RDMA control plane; our SmartNIC study examines how off-path accelerators can support distributed applications. ATC'22 | OSDI'23

• Eurosys Best paper, 2024
• 华为火花奖, 2022
• ACM 中国优秀博士学位论文提名奖 (ACM China Doctoral Dissertation Award), 2021
• ACM ChinaSys 优秀博士学位论文奖 (ACM ChinaSys Doctoral Dissertation Award), 2021
• Honorable Mention of the ACM SIGOPS Dennis M. Ritchie Award, 2021 Link
• Huawei OlympusMons Pioneer Award, 2020 Link
• Microsoft Research Asia Fellowship Award, 2018 Link
• Excellent Bachelor Thesis, 2015

• Program Committee Member
  • 2027: NSDI, OSDI
  • 2026: ASPLOS, EuroSys (Fall)
  • 2025: NSDI, OSDI, CoNEXT, APSys
  • 2024: NSDI, ASPLOS (Fall), CoNEXT (Winter)

Ph.D. Students

• Rongxin Cheng (since 2022; co-advised with Prof. Rong Chen)
• Dingyan Zhang (since 2023)
• Yingyi Hao (since 2024)
• Jiahao Wang (since 2025)

Master's Students

• Tianle Sun (since 2024)
• Jinbo Han (since 2024)
• Yifan Peng (since 2024; co-advised with Prof. Rong Chen)
• Yang Liu (since 2025)
• Shitong Chen (since 2025)
• Kaixi Zhang (since 2026)

• [OSDI'26] Simple is Better: Multiplication May Be All You Need for LLM Request Scheduling.
  Dingyan Zhang, Jinbo Han, Kaixi Zhang, Xingda Wei (corresponding author), Sijie Shen, Chenguang Fang, Wenyuan Yu, Jingren Zhou, Rong Chen.
  USENIX Symposium on Operating Systems Design and Implementation, 2026.
  Paper | Code

• [OSDI'26] SDCs in the Wild: Characterizing and Diagnosing SDC-defective GPUs in Production LLM Training.
  Wenxin Zheng, Bin Xu, Jinyu Gu, Xingda Wei, Haibo Chen, Wenxiao Wang, Yun Zhang, Zuquan Song, Gaohong Liu, Yuchen Nie, Mingcong Han, Zhe Nan, Zhuolin Zheng, Huan Yu, Shuguang Wang, Ziming Zhou, Hang Zhu, Wencong Xiao, Xin Liu.
  USENIX Symposium on Operating Systems Design and Implementation, 2026.

• [PPoPP'26] MetaAttention: A Unified and Performant Attention Framework across Hardware Backends.
  Feiyang Chen, Yu Cheng, Lei Wang, Yuqing Xia, Ziming Miao, Lingxiao Ma, Fan Yang, Jilong Xue, Zhi Yang, Mao Yang, Xingda Wei, Haibo Chen.
  31st ACM SIGPLAN Annual Symposium on Principles and Practice of Parallel Programming, 2026.

• [FAST'26] Fast Cloud Storage for AI Jobs via Grouped I/O API with Transparent Read/Write Optimizations.
  Yingyi Hao, Ting Yao, Xingda Wei (corresponding author), Dingyan Zhang, Tianle Sun, Yiwen Zhang, Zhiyong Fu, Huatao Wu, Rong Chen.
  24th USENIX Conference on File and Storage Technologies, 2026.
  Paper

• [EuroSys'26] KunServe: Parameter-centric Memory Management for Efficient Memory Overloading Handling in LLM Serving.
  Rongxin Cheng, Yuxin Lai, Xingda Wei (corresponding author), Rong Chen, Haibo Chen.
  21st ACM SIGOPS European Conference on Computer Systems, 2026.
  Paper | Code

• [SOSP'25] PhoenixOS: Concurrent OS-level GPU Checkpoint and Restore with Validated Speculation.
  Xingda Wei, Zhuobin Huang (co-first), Tianle Sun, Yingyi Hao, Rong Chen, Mingcong Han, Jinyu Gu, Haibo Chen.
  31st Symposium on Operating Systems Principles, 2025.
  Paper | Code

• [OSDI'25] BlitzScale: Fast and Live Large Model Autoscaling with O(1) Host Caching.
  Dingyan Zhang, Haotian Wang, Yang Liu, Xingda Wei (corresponding author), Yizhou Shan, Rong Chen, Haibo Chen.
  19th USENIX Symposium on Operating Systems Design and Implementation, 2025.
  Paper | Code

• [ATC'25] KVCache Cache in the Wild: Characterizing and Optimizing KVCache Cache at a Large Cloud Provider.
  Jiahao Wang, Jinbo Han, Xingda Wei (corresponding author), Sijie Shen, Dingyan Zhang, Chenguang Fang, Rong Chen, Wenyuan Yu, Haibo Chen.
  2025 USENIX Annual Technical Conference.
  Paper

• [NSDI'25] ODRP: On-Demand Remote Paging with Programmable RDMA.
  Zixuan Wang, Xingda Wei, Jinyu Gu, Hongrui Xie, Rong Chen, and Haibo Chen.
  22nd USENIX Symposium on Networked Systems Design and Implementation, 2025.
  Paper

• [EuroSys'24] Serialization/Deserialization-free State Transfer in Serverless Workflows.
  Fangming Lu, Xingda Wei (corresponding author), Zhuobin Huang, Rong Chen, Mingyu Wu, and Haibo Chen.
  19th ACM SIGOPS European Conference on Computer Systems, Athens, Greece, April 2024. Best Paper Award.
  Paper

• [Preprint] PARALLELGPUOS: A Concurrent OS-level GPU Checkpoint and Restore System using Validated Speculation.
  Zhuobin Huang, Xingda Wei, Yingyi Hao, Rong Chen, Mingcong Han, Jinyu Gu, Haibo Chen.
  2024.

• [Preprint] Characterizing the Dilemma of Performance and Index Size in Billion-Scale Vector Search and Breaking It with Second-Tier Memory.
  Rongxin Cheng, Yifan Peng, Xingda Wei, Hongrui Xie, Rong Chen, Sijie Shen, Haibo Chen.
  2024.
  Preprint

• [OSDI'23] No Provisioned Concurrency: Fast RDMA-codesigned Remote Fork for Serverless Computing.
  Xingda Wei, Fangming Lu, Tianxia Wang, Jinyu Gu, Yuhan Yang, Rong Chen, and Haibo Chen.
  17th USENIX Symposium on Operating Systems Design and Implementation, Boston, MA, US, July 2023.
  Paper | Preprint

• [OSDI'23] Characterizing Off-path SmartNIC for Accelerating Distributed Applications.
  Xingda Wei, Rongxin Cheng, Yuhan Yang, Rong Chen, and Haibo Chen.
  17th USENIX Symposium on Operating Systems Design and Implementation, Boston, MA, US, July 2023.
  Paper

• [ATC'22] KRCORE: A Microsecond-scale RDMA Control Plane for Elastic Computing.
  Xingda Wei, Fangming Lu, Rong Chen, and Haibo Chen.
  2022 USENIX Annual Technical Conference, Carlsbad, CA, US, July 2022.
  Paper | Code / Artifact
  USENIX Badges: Artifacts Available, Artifacts Functional, Results Reproduced.

• [ATC'21] Characterizing and Optimizing Remote Persistent Memory with RDMA and NVM.
  Xingda Wei, Xiating Xie, Rong Chen, Haibo Chen, Binyu Zang.
  2021 USENIX Annual Technical Conference, July 2021.
  Paper

• [NSDI'21] Unifying Timestamp with Transaction Ordering for MVCC with Decentralized Scalar Timestamp.
  Xingda Wei, Rong Chen, Haibo Chen, Zhaoguo Wang, Zhenhan Gong, and Binyu Zang.
  18th USENIX Symposium on Networked Systems Design and Implementation, Boston, MA, US, April 2021.
  Paper | Talk | Code

• [ACM TOS] XStore: A Fast RDMA-based Ordered Key-Value Store with Remote Learned Cache.
  Xingda Wei, Rong Chen, Haibo Chen, and Binyu Zang.
  ACM Transactions on Storage, 17(3), Article 18, August 2021.
  Paper

• [OSDI'20] Fast RDMA-based Ordered Key-Value Store using Remote Learned Cache.
  Xingda Wei, Rong Chen, and Haibo Chen.
  14th USENIX Symposium on Operating Systems Design and Implementation, Banff, Alberta, Canada, November 2020.
  Paper | Talk | Code
  Extended version: ACM TOS.

• [ATC'19] Pragh: Locality-preserving Graph Traversal with Split Live Migration.
  Xiating Xie, Xingda Wei, Rong Chen, and Haibo Chen.
  2019 USENIX Annual Technical Conference, Renton, WA, US, July 2019.
  Paper | Talk

• [OSDI'18] Deconstructing RDMA-enabled Transaction Processing: Hybrid is Better!
  Xingda Wei, Zhiyuan Dong, Rong Chen, and Haibo Chen.
  13th USENIX Symposium on Operating Systems Design and Implementation, Carlsbad, CA, US, October 2018.
  Paper | Slides | Talk | Code

• [ATC'17] Replication-driven Live Reconfiguration for Fast Distributed Transaction Processing.
  Xingda Wei, Sijie Shen, Rong Chen, and Haibo Chen.
  2017 USENIX Annual Technical Conference, Santa Clara, CA, US, July 2017.
  Paper

• [ACM TOCS] Fast In-memory Transaction Processing using RDMA and HTM.
  Haibo Chen, Rong Chen, Xingda Wei, Jiaxin Shi, Yanzhe Chen, Zhaoguo Wang, Binyu Zang, Haibing Guan.
  ACM Transactions on Computer Systems, Vol. 35, No. 1, Article 3, July 2017.
  Paper

• [EuroSys'16] Fast and General Distributed Transactions Using RDMA and HTM.
  Yanzhe Chen, Xingda Wei, Jiaxin Shi, Rong Chen, and Haibo Chen.
  11th ACM European Conference on Computer Systems, London, UK, April 2016.
  Paper | ACM DL

• [SOSP'15] Fast In-memory Transaction Processing using RDMA and HTM.
  Xingda Wei, Jiaxin Shi, Yanzhe Chen, Rong Chen, and Haibo Chen.
  25th ACM Symposium on Operating Systems Principles, Monterey, CA, USA, October 2015.
  Paper | Slides | Poster | ACM DL | Code | The Morning Paper
  Extended version: ACM TOCS.

• Computer system principle, 2022, Co-instructor
• Computer system engineering, 2022, Co-instructor
• Computer system engineering, 2021, Co-instructor
• Advanced Distributed Systems, TA