tagline one



UBDA platform can support various kind of research areas of different departments. It is currently supporting research projects related to Modern Distribution Network, Air Pollution, Vehicular Traffic, Climate System, etc.


Mechanistic Study on Inhibition of Ebselen, AMA and Captopril towards New Delhi Metallo-β-lactamase (NDM-1) by Mass Spectrometry (Dr. Yao Zhongping/ABCT) 

Great efforts have been made to develop NDM-1 inhibitors, and ebselen, aspergillomarasmine A (AMA) and captopril are three compounds that showed significant inhibition activities. However, the inhibition mechanisms of these compounds are still not very clear. In this project, we aim to perform deep investigations on the interactions of NDM-1 with the three inhibitors. Mass spectrometry, an important technique that is complimentary to X-ray crystallography for studying protein-drug interactions, integrated with computer simulation will be applied to investigate various aspects, e.g., protein conformations and dynamics as well as binding stoichiometry and kinetics, of the inhibition mechanisms. The mechanistic insights obtained in this study would be highly beneficial to the implementation and further development of the three NDM-1 inhibitors.  

Mathematical models to help the fight against the COVID-19 pandemic (Dr. He Daihai/AMA)

The project has developed both statistical and differential equation type of models for the transmission of COVID-19. These models stochastically simulated on UBDA. The state-of-the-art iterated filtering method is used to estimate the several key epidemiological parameters of COVID-19, including the basic reproductive number, the serial interval (as a proxy of the generation interval), the dispersion parameter, the relapse rate among recovered and discharged patient, the relative transmission rate of asymptomatic patients, and the epidemic size in Wuhan. The works have been cited more than 600 times according to Google Scholar and caught media attention. We earned a collaborative research grant (HKD450,000) from Alibaba (China) Co. Ltd.

OSP: Overlapping Computation and Communication in Parameter Server for Fast Machine Learning (Prof Guo Song/COMP)

The project is to improve the speed of large-scale model training over Parameter Server architecture. Specifically, training datasets are distributed over multiple workers and a global machine learning model is cooperatively trained with the coordination of a server. In this case, each worker is equipped with a GPU. The project team has run the model training on multiple GPUs for utilizing considerable computing resources. However, with the number of GPUs increasing, the communication overhead of updates aggregation and model synchronization among GPUs becomes the bottleneck. To overcome this bottleneck, the team has proposed Overlapping Synchronization Parallelization (OSP) where each worker exchanges information with the server while simultaneously runs computation on the GPU in a non-stop manner. To evaluate OSP, the team has used multiple GPUs in UBDA to implement distributed experiments on various datasets and machine learning models. The experimental results demonstrate the significant improvement in training efficiency of our method.

Photochemical Air Pollution in Highly Urbanized Subtropical Regions: from Micro Environments to Urban-Terrestrial-Oceanic Interactions (Prof Tao Wang/CEE)

Photochemical air pollution is a pressing environmental problem and is on the top of the agenda of the governments in both HK and mainland China. This comprehensive research is to study the severe photochemical air pollution in HK and the adjacent Pearl River Delta (PRD) region. We aim to understand complex multi-scale processes of photochemical pollution in subtropical high-density urban areas and recommend evidence-based mitigation strategies. This project has been funded by RGC-Theme-based research scheme (T24-504/17-N)

The UBDA platform can support to simulate air quality with three-dimensional regional chemistry and transport models such as WRF-Chem and WRF-CMAQ.

Modern Distribution Networks

Distribution Networks consist of several distribution centers and various demand points dispersed in different regional areas and generate massive amounts of data. The biggest challenge is to use the generated big data for identifying the right number of distribution centers to open the distribution centers.

Air Pollution

As a city infrastructure grows, the level of carbon monoxide, nitrogen dioxide, ground-level ozone, particulates, sulfur dioxide, hydrocarbons, and lead also increases. This increases the risk of strokes, heart disease, lung cancer and pulmonary/acute respiratory diseases, such as asthma.

The big data of a city can be used to calculate the impact of pollution on the human well beings.

Vehicular Traffic

Modern vehicles are capable of reporting location and status information in real time to the vehicle’s control and diagnostic systems. This kind of data can be used to handle several traffic issues, such as congestion, construction of new roads, diversion, etc.  

Climate System

Global climate change and its impact on human life have become one of our era's greatest challenges. The data of changing climate can be used to understand the climate changing patterns.