Research Collaboration and Knowledge Transfer

As outlined in our Research page, we have extensive experience in collaboration with industrial, academic and research institutions worldwide. We have a successful track record with EPSRC, TSB, EU and JISC; and directly with the research arm of institutions.

Therefore, we welcome researchers joining us via a variety of routes – as a research partner, a visiting researcher, a postgraduate student, or an intern.

Postgraduate Research Students

Specific funding or career opportunities will be announced on the school’s main funding opportunities page with accompanying information. Current PhD supervisors are: Professor Jie Xu and Professor Karim Djemame.

There are many opportunities for PhD research within our group. A selection of examples can be seen below, please also feel free to contact the group to discuss your own interests.

  • Cloud Data Analysis
    To date, there has been no widely-used analysis of real Cloud usage. This project seeks to use distributed data processing technology (such as Hadoop) – performing on a large cluster of machines at the University of Leeds – to mine and analyse multiple sets of cloud datacentre data (such as that released by Google) in order to generate meaningful statistics and conclusions that will direct future research into real-world Cloud performance, dependability, and energy usage. These conclusions will then be further tested and analysed on a test cloud developed at the University of Leeds.
  • Risk Management in Clouds
    The objective of the research is to identify the need for risk management at various stages of the cloud service lifecycle (construction, deployment, and operation). It emphasizes on the design of and implementation of various risk models used by different cloud architectural components throughout the service lifecycle. It aims towards optimized service construction, deployment, and execution for Cloud Infrastructures by offering risk management tools to efficiently manage the full life cycle of services. These tools provide simplified construction of services, help make informed deployment and runtime management decisions based on risk assessment models for evaluation of providers, and permit the appropriate establishment of fault tolerance mechanisms.
  • Energy Efficient Cloud Computing
    There has been considerable effort in the Cloud computing industry to reduce costs through the use of energy efficient hardware (cooling, power-saving, etc.) However, an area that has been mostly overlooked is how energy can also be saved through intelligent scheduling of Cloud jobs. This project aims to analyse existing cloud infrastructures and scheduling algorithms in order to improve their energy efficiency; simplex and intercloud environments will be considered, and real hardware monitors will be used to evaluate the effectiveness of the generated schemes.
  • Elastic Real-Time Applications in Cloud Environments
    There has been an increased effort in developing and integrating Real-Time applications into the Cloud environment (streaming, hosted servers for online gaming, etc.) However, designing and constructing Real-Time applications that leverage Cloud elasticity that are capable of enforcing timing constraints in the presence of a turbulent and unpredictable Cloud system environment is not trivial. This project seeks to formalise such a design in order to enable the development of Real-Time Cloud applications that are able to cope and leverage the characteristics of the Cloud computing environment.
  • Definition of a Methodology to Develop High Scalable Applications Using Commercial Cloud Frameworks
    One of the biggest advantages of Cloud Computing over other computational models is that it provides high scalability on demand. However, architecting applications to use this Cloud characteristic is not trivial. It requires a deep understanding of the Cloud computing model and also of the framework supplied by Cloud providers. There is a lack of well-defined methodologies to support system architects in order to exploit the available framework tools to achieve high scalability with the minimum effort. This project seeks to develop and formalise such a method in order to allow consumers to make a more efficient use of resources and improve required performance under growing demand.
  • Making Sense of User Generated Content in a Collaborative Environment
    Social media and cloud platforms are increasingly being used for collaboration. The amount of content and usage data generated in these environments is huge. This poses both challenges and opportunities. These user generated content provided a rich set of captured knowledge for behaviour analyses. Due to the size of data, accuracy and scalability of analysis are some of the issues and require further research. At Leeds, we have embarked on a model-driven approach which utilises semantics as a means of representation for further reasoning to make sense of the underlying patterns. Our research projects have utilised models from Activity Theory, individual and collaborative sense making models, and provenance models depending on the context. Algorithms are being experimented to streamline the pipeline from data/knowledge capture to representation within a collaborative setting.
  • Byzantine Fault-Tolerance in Clouds
    Computing Clouds are typically composed of a large number of physical and virtual machines, capable of supporting a large number of often highly elastic applications from a great number of users. In this environment, it is important to guard against the occurrence of Byzantine faults – arbitrary faults introduced into the system by malicious entities. This project aims to develop technologies to detect Byzantine faults and allow cloud jobs to continue to provide correct service in the presence of such faults. It will involve a rigorous examination of the existing Byzantine Fault-Tolerant (BFT) techniques used in traditional distributed systems and development, deployment and evaluation of a new algorithm to support BFT in real Clouds.

General advice for research degrees applications can be found on the school of computing’s main pages.

Past Visitors

  • Visiting Professor Chaug Guo – NUDT, China
  • Visiting Professor Kaigui Wu – Chongqing University, China
  • Visiting student Shengdong Zhang – NUDT, China

Past Research (Alumni’s Theses)

Our group has had many successful alumni, who have produced their theses here at Leeds. We highlight their success here, while other theses produced in the school can be found under the main school listing.