PhD Opportunity

Big Data: Coping with Data Obesity in Cloud Environments

With the increasing desire among industry and governments to reduce their ICT costs, cloud computing is attracting increasing attention. Coupled with decisions concerning choice of desktop technologies and applications mix, are concerns about maintaining Quality of Service, Data Protection, Network Security and System Resilience and Availability. In certain application scenarios and deployments, the number of host sites can grow significantly, ranging from thousands of users in entire government departments, smart homes, healthcare monitoring and large scale environmental monitoring to large scale commercial applications such as banking and web services for global companies.

In common with the scalability and performance issues for these cloud applications are the challenges associated with data acquisition, aggregation and summarisation that are required in order that as much raw data as possible can be processed locally in-situ at the host location thereby minimising the amount of data that needs to be transmitted over the network for storage and subsequent processing. These problems become compounded with large-scale real-time deployments where the volume of data can grow significantly, based on the frequency of polling, the message passing protocols and the type of data, the latter increasingly multimedia in nature. It follows that the data storage facilities can suffer from data obesity where the pressures of processing in real time to extract knowledge conflict with data currency and usefulness. In addition this leads to inefficient use of constrained network resources.

Research issues to be addressed include: local data acquisition and cleansing, pre-processing to derive sufficient statistics for efficient transmission, aggregation and summarisation protocols, light-weight in-cloud message passing protocols, caching and replication as appropriate, and development of robust methods for data analytics.

The successful applicant should have knowledge of telecommunications protocols, data management and statistical modelling.

Personnel Involved

First Supervisor: Scotney, BW Prof
Second Supervisor: Parr, GP Prof
Third Supervisor: McClean, SI Prof
Collaborator: Professor Nader Azarmi (Visiting Professor University of Ulster and Director of EBTIC-ETISALAT BT Innovation Centre, Abu Dhabi, UAE) This project is part of the EPSRC India-UK Advanced Technology Project (IU-ATC), in association with BT.

Collaboration: This project does not involve collaboration with another establishment

Synopsis:

Return to list of PhD Opportunities