Dr. Kaiyang Liu - Personal Homepage

About Me

I am an Assistant Professor in the Department of Computer Science at Memorial University of Newfoundland (MUN), Canada. Before joining MUN, I worked as a Post Doctoral Fellow at the University of Victoria, Canada with Prof. Jianping Pan. My research primarily focuses on Distributed Cloud/Edge Computing and Storage Networks, Data Center Networks, and Distributed Machine Learning Systems, with a special focus on the analysis and optimization of computation- and data-intensive services. I am a Senior Member of the IEEE (SMIEEE).

Latest News & Highlights

2025.10 🎉 Jinhao Luo has received the 2024-2025 Graduate Research Award, Department of Computer Science, Memorial University of Newfoundland.

2025.07 ✅ Our paper "Performance Analysis of Communication Scheduling Schemes for Distributed Deep Learning" has been accepted to IEEE LCN 2025.

2025.07 ✅ Our paper "Efficient Device Placement for Distributed DNN Training" has been accepted to IEEE ICC 2025.

2025.01 🎉 Hong Wang has received the 2024-2025 Graduate Research Award, Department of Computer Science, Memorial University of Newfoundland.

2024.07 ⭐ Dr. Liu has been elevated to IEEE Senior Member.

2024.07 ✅ Our paper "Sampling-based Multi-job Placement for Heterogeneous Deep Learning Clusters" has been accepted to IEEE Transactions on Parallel and Distributed Systems.

2024.04 🏆 Dr. Liu has received the prestigious 5-Year NSERC Discovery Grants and the Discovery Launch Supplement Award.

2023.09 ✅ Our paper "Sampling-based Caching for Low Latency in Distributed Coded Storage Systems" has been accepted to IEEE Transactions on Services Computing.

Work Experience

Tenure-Track Assistant Professor

September 2023 – Present

Department of Computer Science, Memorial University of Newfoundland, St John's, NL, Canada

Visiting Professor

May 2023 – August 2023

Department of Computer Science, University of Victoria, Victoria, BC, Canada

Postdoctoral Fellow

November 2019 – April 2023

Department of Computer Science, University of Victoria, Victoria, BC, Canada

Education

Ph.D. in Traffic Information Engineering and Control

September 2014 – June 2019

Central South University, Changsha, Hunan, China

Thesis: Efficient computing task scheduling and data placement schemes for big data in cloud data centers (Supervisor: Prof. Jun Peng)

M.Sc. in Information and Communication Engineering

September 2012 – August 2014

Central South University, Changsha, Hunan, China

Research Program: Coalition formation and pricing mechanism for resource sharing in cloud computing (Supervisor: Prof. Jun Peng)

B.Eng. in Communication Engineering

September 2008 – June 2012

Central South University, Changsha, Hunan, China

Thesis: Heterogeneous architecture design of wireless sensor/mesh networks and its hardware-in-the-loop implementation (Supervisor: Prof. Jun Peng)

Network Intelligence for Next-generation Cloud Computing

Next-generation cloud computing and Artificial Intelligence (AI) are jointly reshaping the digital ecosystem. AI is driving cloud platforms toward intelligent, data-centric architectures, while cloud computing provides the scalability and efficiency essential for modern AI workloads. This interdependence creates both opportunities and challenges. AI techniques developed for vision and language tasks often incur high overhead and unreliable performance when applied to network optimization. Conversely, current cloud infrastructures were designed for general-purpose resource sharing and lack AI-aware network and protocol optimizations. Advancing cloud network architectures to meet the stringent requirements of AI applications is therefore a key research problem.

Application-aware Cloud Networking

Contemporary cloud networks rely on a general-purpose, best-effort design that lacks performance guarantees. This model is increasingly inadequate for delay-sensitive services and specialized workloads such as large-scale AI/ML and GPU-based tasks. Studies show significant performance variability—particularly in tail latency—across current cloud providers. These challenges motivate an application-aware cloud networking approach. We focus on redesigning resource management, scheduling, and congestion detection, and on optimizing AI/ML workloads through integration with communication libraries like NVIDIA NCCL.

Low-Earth-Orbit Satellite (LEOS)-based Space Cloud/Edge Computing

LEOS-based space edge computing has recently gained significant attention, driven by initiatives such as Lumen Orbit and AI Sweden Space Edge. Hosting data storage and processing in space offers a compelling vision of scalable, energy-efficient, and low-latency computing for applications ranging from finance to real-time communication and remote sensing. My prior work centers on distributed optimization and AI techniques across cloud/edge systems, data center networks, and satellite communication networks to meet the demands of data-intensive applications. Key contributions include efficient task scheduling, data placement, and caching mechanisms designed for high performance and scalability. These solutions must be reexamined to ensure low latency, high throughput, and resilience under dynamic traffic and satellite conditions. Establishing a coast-to-coast-to-coast LEO satellite network testbed will further illuminate the challenges and opportunities of future space-based data centers.

Instructor - Department of Computer Science, Memorial University of Newfoundland

COMP 4759 - Computer Networks (Fall 2023, Fall 2024, Fall 2025)

COMP 6777 - Mobile Ad hoc Networking (Fall 2023)

COMP 6910 - Services Computing, Semantic Web and Cloud Computing (Spring 2025, Spring 2026)

COMP 6917 - Complex Networks (Fall 2025)

Previous Teaching/Guest Lectures - Department of Computer Science, University of Victoria

CSC 466/579 - Advanced Computer Networks (Spring 2020, Spring 2021, Fall 2022)

CSC 360 - Operating Systems (Spring 2022)

CSC 361 - Computer Communication and Networks (Fall 2022)

Selected Refereed Journal Papers

Sampling-based multi-job placement for heterogeneous deep learning clusters

Kaiyang Liu, Jingrong Wang, Zhiming Huang, Jianping Pan. IEEE Transactions on Parallel and Distributed Systems, vol. 35, no. 6, pp. 874-888, 2024.

Sampling-based caching for low latency in distributed coded storage systems

Kaiyang Liu, Jingrong Wang, Heng Li, Jun Peng, Jianping Pan. IEEE Transactions on Services Computing, vol. 16, no. 6, pp. 4275-4287, 2023.

Adaptive and scalable caching for low latency in distributed coded storage systems

Kaiyang Liu, Jun Peng, Jingrong Wang, Zhiwu Huang, Jianping Pan. IEEE Transactions on Cloud Computing, vol. 11, no. 2, pp. 1840-1853, 2023.

Optimal caching for low latency in distributed coded storage systems

Kaiyang Liu, Jun Peng, Jingrong Wang, Jianping Pan. IEEE/ACM Transactions on Networking, vol. 30, no. 3, pp. 1132-1145, 2022.

A learning-based data placement framework for low latency in data center networks

Kaiyang Liu, Jun Peng, Jingrong Wang, Boyang Yu, Zhuofan Liao, Zhiwu Huang, Jianping Pan. IEEE Transactions on Cloud Computing, vol. 10, no. 1, pp. 146-157, 2022.

An instance reservation framework for cost effective services in geo-distributed data centers

Kaiyang Liu, Jun Peng, Boyang Yu, Weirong Liu, Zhiwu Huang, Jianping Pan. IEEE Transactions on Services Computing, vol. 14, no. 2, pp. 356-370, 2021.

Scalable and adaptive data replica placement for geo-distributed cloud storages

Kaiyang Liu, Jun Peng, Jingrong Wang, Weirong Liu, Zhiwu Huang, Jianping Pan. IEEE Transactions on Parallel and Distributed Systems, vol. 31, no. 7, pp. 1575-1587, 2020.

An active mobile charging and data collection scheme for clustered sensor networks

Kaiyang Liu, Jun Peng, Liang He, Jianping Pan, Shuo Li, Ming Ling, Zhiwu Huang. IEEE Transactions on Vehicular Technology, vol. 68, no. 5, pp. 5100-5113, 2019.

Multi-device task offloading with time-constraints for energy efficiency in mobile cloud computing

Kaiyang Liu, Jun Peng, Heng Li, Xiaoyong Zhang, Weirong Liu. Future Generation Computer Systems, vol. 64, pp. 1-14, 2016.

Online UAV-mounted edge server dispatching for mobile-to-mobile edge computing

Jingrong Wang, Kaiyang Liu, Jianping Pan. IEEE Internet of Things Journal, vol. 7, no. 2, pp. 1375-1386, 2020.

Selected Refereed Conference Papers

Performance analysis of communication scheduling schemes for distributed deep learning

Jinhao Luo, Hong Wang, Jingrong Wang, Adrian Fiech, Kaiyang Liu. IEEE Conference on Local Computer Networks, Accepted for publication, 2025.

Efficient device placement for distributed DNN training

Hong Wang, Jinhao Luo, Kaiyang Liu, Qiang Ye. IEEE International Conference on Communications, Accepted for publication, 2025.

End-to-end congestion control as learning for unknown games with Bandit feedback

Zhiming Huang, Kaiyang Liu, Jianping Pan. IEEE International Conference on Distributed Computing Systems, pp. 327-338, 2023.

Energy-aware inter-data center VM migration over elastic optical networks

Fatima S. Amri, Zhiming Huang, Kaiyang Liu, Jianping Pan. IEEE Global Communications Conference, pp. 5421-5426, 2023.

Learning-based adaptive data placement for low latency in data center networks

Kaiyang Liu, Jingrong Wang, Zhuofan Liao, Boyang Yu, Jianping Pan. IEEE Conference on Local Computer Networks, pp. 142-149, 2018. (Best Paper Award Candidate)

A combinatorial optimization for energy-efficient mobile cloud offloading over cellular networks

Kaiyang Liu, Jun Peng, Xiaoyong Zhang, Zhiwu Huang. IEEE Global Communications Conference, pp. 1-6, 2016.

Dynamic resource reservation via broker federation in cloud service: A fine-grained heuristic-based approach

Kaiyang Liu, Jun Peng, Weirong Liu, Pingping Yao, Zhiwu Huang. IEEE Global Communications Conference, pp. 2338-2343, 2014.

View All Publications on Google Scholar »

Current Students

Jinhao Luo (Ph.D., 2025.09 - Present; M.Sc., 2024.09 - 2025.07)
Research Topic: Application-aware Cloud Networking
Hong Wang (Ph.D., 2025.09 - Present; M.Sc., 2023.09 - 2025.07)
Research Topic: Distributed Machine Learning Systems
Jing Jie Tan (Visiting Ph.D., 2025.09 - 2026.01)
Universiti Tunku Abdul Rahman, Malaysia

Canada-ASEAN Scholarships and Educational Exchanges for Development

Research Topic: Privacy-Preserving Machine Learning

Past Students

Vahidreza Niazmand (M.Sc., Graduated 2024.12)
Thesis: Joint task offloading, DNN pruning, and computing resource allocation for fault detection with dynamic constraints in industrial IoT
Current Role: Senior Application Developer at Sobeys in Mississauga, Canada
Pritom Anthony Rozario (Undergraduate Honors Program, 2024.09 - 2025.08)
Research Project: Performance Evaluation of Topology-aware Distributed Learning in Data Center Networks

To Prospective Ph.D. and M.Sc. Students

I currently have open M.Sc./Ph.D. positions. Highly self-motivated students with a strong background in Computer Science, Computer Engineering, or a related field are encouraged to apply. Areas of research include Network Intelligence for Next-generation Cloud Computing, Protocols Design for Advanced Networking, and Distributed Machine Learning. If interested, please send me by email your CV, transcripts, sample publications (if any), and TOFEL/IELTS test results, with the email subject line "Prospective [M.Sc./Ph.D.] Student - [Your Name]". Admitted Ph.D./M.Sc. students will be provided with full financial support. Admission to the Ph.D. or the M.Sc. program is mainly granted for the Fall intake semesters (application deadline: December 1st). Please refer to the application guides on the Faculty of Graduate Studies website for M.Sc. and Ph.D. programs. Visiting scholars/students are also welcomed worldwide.