Faculty

Xingda Wei
Tenured Track Associate Professor

Email:wxdwfc@sjtu.edu.cn

Institute:Institute of Parallel And Distributed Systems

MainPage:https://ipads.se.sjtu.edu.cn/pub/members/xingda_wei

Brief Introduction

My research focuses on building system infrastructures to systematically bring better performance and resilience to emerging applications like AI training, inference, and agents.

 

My recent works target topics like: Scalable AI infrastructure via fast autoscaling and dynamic resource management. Representative work: BlitzScale (OSDI'25), PhoenixOS (SOSP'25), KunServe (EuroSys'26). Efficient system to harness computing resources. Representative work: LMMetric (OSDI'26), SpecActor (arXiv) and fast data planes for AI workloads. Representative work: AITurbo (FAST'26), KVCache cache study (ATC'25). 


Many of my recent works have been or is being deployed on large AI service providers like Huawei, Alibaba and Bytedance Seed. 


For more information, please check: https://ipads.se.sjtu.edu.cn/pub/members/xingda_wei


Teaching Assignment

- Computer system engineering

- Computer system principles 

Publications

Selected publications: 

  • [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. The 16th USENIX Symposium on Operating Systems Design and Implementation.

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

  • [OSDI'25] Fast and Live Model Auto Scaling with O(1) Host Caching. Dingyan Zhang, Haotian Wang, Yang Liu, Xingda Wei (Corresponding author), Yizhou Shan, Rong Chen, Haibo Chen. The 19th USENIX Symposium on Operating Systems Design and Implementation.

  • [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.

  • [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.

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

  • [OSDI'23] No Provisioned Concurrency: Fast RDMA-codesigned Remote Fork for Serverless ComputingXingda 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.

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


Awards

  • 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 

  • Huawei OlympusMons Pioneer Award, 2020 

  • Microsoft Research Asia Fellowship Award, 2018 

  • Excellent Bachelor Thesis, 2015


Academic Service

Program Committee Member

  • 2027: NSDI 

  • 2026: ASPLOS, EuroSys 

  • 2025: NSDI, OSDI, CoNEXT, APSys

  • 2024: NSDI, ASPLOS (Fall), CoNEXT (Winter)

  • 2023: SysDW, Chinasys (Fall)

  • 2022: EuroDW