| EPSRC Reference: |
GR/T18615/01 |
| Title: |
Whole System Modelling Of Large-Scale Communication Networks For What-If Evaluation |
| Principal Investigator: |
Professor JM Pitts |
| Other Investigators: |
|
| Researcher Co-investigators: |
|
| Project Partners: |
|
| Department: |
Sch of Electronic Eng & Computer Science |
| Organisation: |
Queen Mary, University of London |
| Scheme: |
Standard Research |
| Starts: |
01 October 2004 |
Ends: |
30 September 2007 |
Value (£): |
237,237
|
| EPSRC Research Topic Classifications: |
| Information and communication technologies: ICT Networks and Distributed Systems |
|
|
| EPSRC Industrial Sector Classifications: |
|
| Related Grants: |
|
| Panel History: |
|
|
Summary |
Common simulation platforms have played a key role in the design and deployment of wide-area Internet protocols. These protocols are vital to the exchange of electronic information - the life-blood of the modem knowledge-based economy.
This project aims to develop new methods for supporting the simulation of large-scale communications networks by using national supercomputing services. Many areas of computational science and engineering research are increasingly being focused on the simulation of whole systems, rather than just system components. This project aims to create a new user community that will harness these high-end computing facilities (a) for researching the next generation of networking services and applications, and (b) for what-if evaluation of carrier-scale (i.e. whole system) mobile and broadband network infrastructure. '
The research will be based around open, extensible simulation technologies (ns-from the Internet community; HLA-from the defence sector) to encourage speedy take-up by other networking research teams. The project will also address the significant methodological challenges inherent in large-scale network simulation. Shared network infrastructures are known to exhibit complex, highly variable and non -linear patterns of behaviour. The methodological challenges relate to the scale and complexity of network scenarios, the measurement, analysis and visualisation of results, and the reproducibility and validation of research outcomes for large-scale scenarios.
The research will be undertaken by the Network and Service Assurance Laboratory, part of the Communications Research Group at Queen Mary, University of London, in conjunction with national supercomputing services at the University of Manchester.
|
| Final Report Summary |
This project was proposed in response to a call for development projects to introduce new users to the national supercomputing services. The aim of the call was to encourage UK researchers outside the traditional high-end computing (HEC) community to try out the capability offered by these services and see if they could offer opportunities for new scientific or engineering insight. Many areas of computational science and engineering research are increasingly being focused on the simulation of whole systems, rather than just system components. This project's primary aims were to develop new methods for supporting computer simulations of large-scale communications networks, and to foster a new HEC user community of communications network researchers.
Simulation investigations form a vital part of networking research and have been used by the research community to support innovation in many areas of communication networks. These research advances have been crucial in the development of the global Internet, and internet technology, both of which are vital components of ubiquitous and critical services in the modern Digital Economy.
The international networking community has, over the past decade, developed and supported a number of common simulation platforms to help facilitate the reproducibility of research results. This is a vital component of engineering endeavour, enabling the efficient evaluation and comparison of, for example, alternative Internet protocols to carry real-time services such as video conferencing. However, to enable HEC to be adopted as a general purpose tool for networking research, these developments need to be incorporated into an overall methodological framework that can support exploration of large-scale state spaces in network modelling.
This project has taken a number of important steps towards that goal. It has developed a suite of whole-system benchmark scenarios for enterprise networks, wireless cities, and ad-hoc networks (e.g. to support emergency response). These have helped the project to focus on the scalability issues, particularly as they affect simulation methodology. As a result, the project has made key advances in four areas: (1) whole-of-network Quality of Experience measures; (2) large-scale simulation granularity; (3) accelerated models of Internet priority mechanisms; and (4) ways to model and optimize network topologies at different levels of abstraction. These advances enable researchers not only to organize and run faster simulations more efficiently, but to analyze and interpret results more effectively.
As part of our drive to foster a new HEC user community, the project team at Queen Mary, University of London has been collaborating with BT's Mobility Research Centre to apply, in particular, the whole-of-network measurement methodology to the behaviour of wireless cities scenarios. Many people now are familiar with wireless broadband (known as WiFi) to connect their home computer to the Internet without the hassle of wires. WiFi hot-spots (also called "access points") provide the same service, but in public places such as cafs, hotels and airports. Wireless cities enable Internet access using the same WiFi technology, but without having to search for a hot-spot. The challenge of providing blanket WiFi coverage is that the radio signals from neighbouring access points can interfere if they are too close to each other, and this can slow down an Internet connection. Ongoing collaborative work is investigating how this interference affects the quality of the Internet connection. The project's methods are helping researchers to cope with the huge number of different patterns of behaviour that can arise, so that they can have confidence in the results obtained from the supercomputer models, and to evaluate new techniques to minimize interference and maximize quality.
|
| Further Information: |
|
| Organisation Website: |
http://www.qmul.ac.uk |