Contact
Email: luozywh [at] outlook.com
Office: Room C509, School of Cyber Science and Engineering
Wuhan University, Wuhan, China
About
I am a professor in the School of Cyber Science and Engineering at Wuhan University. Previously, I was a postdoctoral researcher and then a research scientist at The Ohio State University, working with Ness B. Shroff and Jia (Kevin) Liu.
I received my Ph.D. in Computer Science from The University of Hong Kong, advised by Chuan Wu, and my B.E. from Wuhan University, advised by Zongpeng Li.
Research
My research focuses on optimizing distributed systems, with an emphasis on efficient resource scheduling. Recently, I have been working on the analysis and optimization of distributed machine learning systems.
Publications
Google Scholar ↗2026
Toward WAN-Aware LLM Training Across Heterogeneous, Geo-Distributed Sites
in Proc. of the 3rd Workshop on Networks for AI Computing (NAIC), co-located with ACM SIGCOMM 2026, Denver, CO, USA, August 17, 2026, to appear.
2025
PipeTGL: (Near) Zero Bubble Memory-based Temporal Graph Neural Network Training via Pipeline Optimization
in Proceedings of the VLDB Endowment, vol. 18, no. 8, pp. 2722-2734, 2025.
SIEVE: A Scalable and General Purpose Data Filtering System for Large Language Models
in Proc. of ICML 2025 Workshop on DataWorld: Unifying Data Curation Frameworks Across Domains, Vancouver, Canada, July 19, 2025.
Finite-Time Global Optimality Convergence in Deep Neural Actor-Critic Methods for Decentralized Multi-Agent Reinforcement Learning
in ICML, PMLR, vol. 267, pp. 75853-75877, Vancouver, Canada, July 13-19, 2025.
Prediction-Assisted Online Distributed Deep Learning Workload Scheduling in GPU Clusters
in IEEE INFOCOM, pp. 1-10, London, United Kingdom, May 19-22, 2025.
2024
Optimizing Task Placement and Online Scheduling for Distributed GNN Training Acceleration in Heterogeneous Systems
in IEEE/ACM Transactions on Networking, vol. 32, no. 5, pp. 3715-3729, October 2024.
Expediting Distributed GNN Training with Feature-only Partition and Optimized Communication Planning
in IEEE INFOCOM, Vancouver, Canada, May 20-23, 2024.
DiffusionPipe: Training Large Diffusion Models with Efficient Pipelines
in the Seventh Conference on Machine Learning and Systems (MLSys), Santa Clara, USA, May 13-16, 2024.
2023
Two-level Graph Caching for Expediting Distributed GNN Training
in IEEE INFOCOM, New York, USA, May 17-20, 2023.
2022
Optimizing Task Placement and Online Scheduling for Distributed GNN Training Acceleration
in IEEE INFOCOM, online, May 2-5, 2022.
Efficient Pipeline Planning for Expedited Distributed DNN Training
in IEEE INFOCOM, online, May 2-5, 2022.
2021
Joint Model and Data Adaptation for Cloud Inference Serving
in the 42th IEEE Real-time Systems Symposuim (RTSS), Dortmund, Germany, December 7-10, 2021.
2020
Fast Training of Deep Learning Models over Multiple GPUs
in ACM/IFIP Middleware, Delft, The Netherlands, December 7-11, 2020.
Optimizing Distributed Training Deployment in Heterogeneous GPU Clusters
in ACM CoNEXT, Barcelona, Spain, December 1-4, 2020.
An Online Algorithm for VNF Service Chain Scaling in Datacenters
in IEEE/ACM Transactions on Networking, vol. 28, No. 3, pp. 1061-1073, June 2020.
2019
Scaling Geo-distributed Network Function Chains: A Prediction and Learning Framework
in IEEE Journal on Selected Areas in Communications (JSAC) Special issue on Network Softwarization & Enablers, vol. 37, no. 8, pp. 1838-1850, August 2019.
2018
Online Cloud Resource Allocation and Pricing with Server Speed Scaling
in IEEE ICC 2018, Kansas City, MO, USA, May 20-24, 2018.