About

Hey, I am Ziming! I am a second-year PhD student at UC Berkeley, working with Prof. Ion Stoica and Prof. Scott Shenker. I am affiliated with Sky Computing and NetSys Lab. I completed my undergrad from Yale University, with double major in Computer Science and Philosophy.

Earlier, I worked with Atul Adya on distributed systems and caching. During undergrad, I was advised by Prof. Anurag Khandelwal, Prof. Dragomir Radev at Yale, and Prof. Rashmi Vinayak at CMU. I am fortunate to work with collaborators from Microsoft Research, NetApp, Facebook AI, Salesforce Research, Twitter, and Databricks. I was awarded Berkeley College of Engineering Fellowship (2023), Outstanding Undergraduate Researcher Award (2023) Runner-Up, Computer Science Prize (2023), and Rosenfeld Science Scholars Fellowship (2021). My works have received best-paper awards at EuroSys and NSDI.

Research

My research areas are broadly in systems and networking, in the context of cloud, data and AI.

• UCCL, a fast and extensible GPU communication library that supports heterogeneous GPU and networking vendors.
• Clink, a consistent distributed caching architecture for high-volume search and AI serving.
• SkyServe, serving AI models across regions and clouds over Spot and On-Demand GPUs.
• SkyPilot, a framework for running LLMs, AI, and batch jobs on any cloud.
• Ray Data, a framework for efficient execution of ML training and inference pipelines over hetereogenous resources.

Some earlier projects:

• Trinity, a distributed data store that achieves both fast multi-attribute queries and storage efficiency.
• I worked on NLP in Yale LILY, on retrieval-based generation and table-based question answering.

Education

• Ph.D. in Computer Science, UC Berkeley, 2023 - Present
• B.S. in Computer Science and Philosophy, Yale University, 2019 - 2023

Selected Publications

SkyServe: Serving AI Models across Regions and Clouds with Spot Instances
Ziming Mao*, Tian Xia*, Zhanghao Wu, Wei-Lin Chiang, Tyler Griggs, Romil Bhardwaj, Zongheng Yang, Scott Shenker, Ion Stoica
EuroSys 2025
[Paper] [Code]

Spirit: Fairness for Interdependent Cache and Bandwidth Resources.
Seung-seob Lee, Jachym Putta, Ziming Mao, Anurag Khandelwal
SOSP 2025

Revisiting Cache Freshness for Emerging Real-Time Applications
Ziming Mao, Rishabh Iyer, Scott Shenker, Ion Stoica
HotNets 2024
[Paper]

Trinity: A Fast Compressed Multi-attribute Data Store
Ziming Mao, Anurag Khandelwal, Kiran Srinivasan
EuroSys 2024, Best Student Paper Award
[Paper] [Code]

DYLE: Dynamic Latent Extraction for Abstractive Long-Input Summarization
Ziming Mao*, Chen Henry Wu*, Ansong Ni, Yusen Zhang, Rui Zhang, Tao Yu, Budhaditya Deb, Chenguang Zhu, Ahmed Hassan Awadallah, Dragomir R. Radev
ACL 2022
[Paper] [Code]

Can’t Be Late: Optimizing Spot Instance Savings under Deadlines
Zhanghao Wu, Wei-Lin Chiang, Ziming Mao, Zongheng Yang, Eric Friedman, Scott Shenker, Ion Stoica
NSDI 2024, Outstanding Paper Award
[Paper] [Code]

Additional Publications

LEANN: A Low-Storage Vector Index
Yichuan Wang, Shu Liu, Zhifei Li, Yongji Wu, Ziming Mao, Yilong Zhao, Xiao Yan, Zhiying Xu, Yang Zhou, Ion Stoica, Sewon Min, Matei Zaharia, Joseph E. Gonzalez
ICML 2025 VecDB Workshop
[Paper]

An Extensible Software Transport Layer for GPU Networking
Yang Zhou, Zhongjie Chen, Ziming Mao, ChonLam Lao, Shuo Yang, Pravein Govindan Kannan, Jiaqi Gao, Yilong Zhao, Yongji Wu, Kaichao You, Fengyuan Ren, Zhiying Xu, Costin Raiciu, Ion Stoica
ArXiv 2025
[Paper]

SkyLB: A Locality-Aware Cross-Region Load Balancer for LLM Inference
Tian Xia, Ziming Mao, Jamison Kerney, Ethan J. Jackson, Zhifei Li, Jiarong Xing, Scott Shenker, Ion Stoica
ArXiv 2025
[Paper]

Locality-aware Fair Scheduling in LLM Serving
Shiyi Cao, Yichuan Wang, Ziming Mao, Pin-Lun Hsu, Liangsheng Yin, Tian Xia, Dacheng Li, Shu Liu, Yineng Zhang, Yang Zhou, Ying Sheng, Joseph Gonzalez, Ion Stoica
ArXiv 2024
[Paper]

The Streaming Batch Model for Efficient and Fault-Tolerant Heterogeneous Execution
Frank Sifei Luan, Ziming Mao, Ron Yifeng Wang, Charlotte Lin, Amog Kamsetty, Hao Chen, Cheng Su, Balaji Veeramani, Scott Lee, SangBin Cho, Clark Zinzow, Eric Liang, Ion Stoica, Stephanie Wang
ArXiv 2024
[Paper] [Code]

GL-Cache: Group-level Learning for Efficient and High-Performance Caching
Juncheng Yang, Ziming Mao, Rashmi Vinayak, Yao Yue
FAST 2023
[Paper] [Code]

Pie: Pooling CPU Memory for LLM Inference.
Yi Xu, Ziming Mao, Xiangxi Mo, Shu Liu, Ion Stoica
ArXiv 2024
[Paper]

Summ^N: A Multi-Stage Summarization Framework for Long Input Dialogues and Documents
Yusen Zhang, Ansong Ni, Ziming Mao, Chen Henry Wu, Chenguang Zhu, Budhaditya Deb, Ahmed Hassan Awadallah, Dragomir R. Radev, Rui Zhang
ACL 2022
[Paper] [Code]

FeTaQA: Free-form Table Question Answering
Linyong Nan, Chiachun Hsieh, Ziming Mao, Xi Victoria Lin, Neha Verma, Rui Zhang, Wojciech Kryściński, Nick Schoelkopf, Riley Kong, Xiangru Tang, Murori Mutuma, Ben Rosand, Isabel Trindade, Renusree Bandaru, Jacob Cunningham, Caiming Xiong, Dragomir Radev
TACL 2022
[Paper] [Code]

Investigating Crowdsourcing Protocols for Evaluating the Factual Consistency of Summaries
Xiangru Tang, Alexander Fabbri, Haoran Li, Ziming Mao, Griffin Adams, Borui Wang, Asli Celikyilmaz, Yashar Mehdad, Dragomir Radev
NAACL 2022
[Paper]

Enhancing fouling mitigation of submerged flat-sheet membranes by vibrating 3D-spacers
Yong Zen Tan, Ziming Mao, Yanjun Zhang, Wen See Tan, Tzyy Haur Chong, Bing Wu, Jia Wei Chew
Sep. Purif. Technol. 2019
[Paper]

Spacer vibration for fouling control of submerged flat sheet membranes
Bing Wu, Yanjun Zhang, Ziming Mao, Wen See Tan, Yong Zen Tan, Jia Wei Chew, Tzyy Haur Chong, Anthony G Fane
Sep. Purif. Technol. 2019
[Paper]

Services

• Artifact Evaluation Committee: EuroSys 25’, FAST 25’
• Organizer: Sky Systems Seminar, UC Berkeley

Miscellaneous

• I grew up in Ningbo, China and studied in Singapore at Hwa Chong Institution.
• I play badminton every week! 🏸
• I will be down for any stargazing trips 🔭 :）