Distributed Systems and Networking (DSAN)
The DSAN research group is conducting cutting edge research on the architectures, protocols and algorithms for distributed systems and networks. The research topics include high-performance GPU computing, cloud computing and data center networking, Internet computing, Internet of Things (IoT) and sensor networks, wireless and mobile networks, and multimedia communication.
Faculty Involved:
Funded Research and Consultancy Projects in the Past Few Years:
| Decentralized Online Optimization with Hard Constraints: Adapting to Dynamics of Training Data, Wireless Channel, and Network Topology |
 | Staff |
Prof. WANG, Juncheng |
| Objectives |
- Explore constrained online optimization theories for performance guarantees despite delayed information.
- Design federated online learning frameworks to optimize learning performance under dynamic resource constraints, with delayed data and channel information.
- Develop decentralized wireless online learning algorithms, to facilitate collaboration among neighboring devices with time-varying communication and computation power, in the presence of delayed data, channel, and topology information.
|
| Grant |
National Natural Science Foundation of China (NSFC) Research Fund |
| Distributed Machine Learning at Wireless Edge Networks: Joint Online Optimization of Computation and Communication |
 | Staff |
Prof. WANG, Juncheng |
| Objectives |
- Address a "when" to optimize problem, by designing an information freshness based online scheduling algorithm, to optimize the performance of online distributed learning with asynchronous computation under limited communication resources;
- Address a "what" to optimize problem, by exploring new distributed constrained online optimization theory, to jointly consider the improvement in machine learning model training and the costs in model aggregation, under dynamic computation and communication systems;
More
- Address a "where" to optimize problem, by developing an online distributed robust learning framework to effectively collaborate wireless devices, edge hosts, and cloud servers over multi-level hierarchical networks, for trading off computation and communication performance.
|
| Grant |
Early Career Scheme (ECS) |
| Optimizing and Accelerating Graph Neural Networks for LargeScale Irregular IoT Sensor Data on Chinese NPU Devices |
 | Staff |
Prof. ZHOU, Amelie Chi |
| Objectives |
- Design a graph neural network model that incorporates patching techniques to effectively handle irregularly sampled sensor time series data.
-
Optimize the gradient communication efficiency in distributed GNN training on Chinese NPUs, thereby improving overall training speed and hardware resource utilization.
-
Develop a low-power, lightweight graph neural network model tailored for Chinese IoT NPUs (e.g., Huawei Ascend), and validate its efficient distributed training in resource-constrained edge environments.
|
| Grant |
Guangdong and Hong Kong Universities “1+1+1”Joint Research Collaboration Scheme |
