Qinghao Hu (胡擎昊)

I am a Postdoctoral Associate working with Prof. Song Han at MIT. I also work closely with Prof. Ana Klimović from ETH. I obtained my Ph.D. degree from NTU in 2023, advised by Prof. Tianwei Zhang and Prof. Yonggang Wen. Before that, I received my M.Sc. degree from National University of Singapore in 2020 and my B.Eng. degree from Zhejiang University in 2018.

Email: [email protected]
Address:
Room344, MIT Building 38
50 Vassar Street, Cambridge, MA 02139

Research Interest
Systems for Large Models
Datacenter Management and Scheduling
Machine Learning for Systems
Publications
[NSDI '24]
Characterization of Large Language Model Development in the Datacenter
Qinghao Hu*, Zhisheng Ye*, Zerui Wang*, Guoteng Wang, Meng Zhang, Qiaoling Chen, Peng Sun, Dahua Lin, Xiaolin Wang, Yingwei Luo, Yonggang Wen, Tianwei Zhang
Paper / System / Data / USENIX ;login:
[SC '24]
TorchGT: A Holistic System for Large-scale Graph Transformer Training
Meng Zhang*, Jie Sun*, Qinghao Hu, Peng Sun, Zeke Wang, Yonggang Wen, Tianwei Zhang
Paper / Code / Artifact Badges: Available Functional Reproduced
[ICDE '24]
Sylvie: 3D-adaptive and Universal System for Large-scale Graph Neural Network Training
Meng Zhang, Qinghao Hu, Cheng Wan, Haozhao Wang, Peng Sun, Yonggang Wen, Tianwei Zhang
Paper / Code
[CSUR '24]
Deep Learning Workload Scheduling in GPU Datacenters: A Survey
Zhisheng Ye*, Wei Gao*, Qinghao Hu*, Peng Sun, Xiaolin Wang, Yingwei Luo, Tianwei Zhang, Yonggang Wen
Paper / Awesome List / ACM Computing Surveys
[OSDI '23]
Hydro: Surrogate-Based Hyperparameter Tuning Service in Datacenters
Qinghao Hu, Zhisheng Ye, Meng Zhang, Qiaoling Chen, Peng Sun, Yonggang Wen, Tianwei Zhang
Paper / Code / Artifact Badges: Available Functional Reproduced
[ASPLOS '23]
Lucid: A Non-Intrusive, Scalable and Interpretable Scheduler for Deep Learning Training Jobs
Qinghao Hu*, Meng Zhang*, Peng Sun, Yonggang Wen, Tianwei Zhang
Paper / Code / Artifact Badges: Available Functional Reproduced
Distinguished Paper Award
[ATC '22]
Primo: Practical Learning-Augmented Systems with Interpretable Models
Qinghao Hu, Harsha Nori, Peng Sun, Yonggang Wen, Tianwei Zhang
Paper / Code / Artifact Badges: Available Functional Reproduced
[SC '21]
Characterization and Prediction of Deep Learning Workloads in Large-Scale GPU Datacenters
Qinghao Hu, Peng Sun, Shengen Yan, Yonggang Wen, Tianwei Zhang
Paper / Code / Data / Artifact Badges: Available Functional Reproduced
Preprint
[arXiv '24]
LoongTrain: Efficient Training of Long-Sequence LLMs with Head-Context Parallelism
Diandian Gu, Peng Sun, Qinghao Hu, Ting Huang, Xun Chen, Yingtong Xiong, Guoteng Wang, Qiaoling Chen, Shangchun Zhao, Jiarui Fang, Yonggang Wen, Tianwei Zhang, Xin Jin, Xuanzhe Liu
Paper / Submitted to a Conference
[arXiv '24]
InternEvo: Efficient Long-Sequence Large Language Model Training via Hybrid Parallelism and Redundant Sharding
Qiaoling Chen, Diandian Gu, Guoteng Wang, Xun Chen, Yingtong Xiong, Ting Huang, Qinghao Hu, Xin Jin, Yonggang Wen, Tianwei Zhang, Peng Sun Liu
Paper / Submitted to a Conference
[arXiv '23]
AMSP: Super-Scaling LLM Training via Advanced Model States Partitioning
Qiaoling Chen, Qinghao Hu, Zhisheng Ye, Guoteng Wang, Peng Sun, Yonggang Wen, Tianwei Zhang
Paper / Submitted to a Conference
Professional Services
[EuroSys '24] Shadow Committee Member
[EuroSys '23] Shadow Committee Member
[OSDI '22] AE Committee Member
[ATC '22] AE Committee Member
[EuroSys '22] AE Committee Member
[SOSP '21] AE Committee Member

Awards
ML and Systems Rising Stars 2024
Outstanding PhD Thesis Award 2024
National Scholarship for Outstanding Graduates 2024
Google PhD Fellowship 2023
Distinguished Paper Award of ASPLOS 2023
Youth Outstanding Paper Award of WAIC 2023
Student Travel Grant of OSDI 2023
Best Undergraduate Thesis Award 2018
Outstanding Graduates of Zhejiang University 2018