This is a talk about TCP throughput and how it relates to network parameters such as delay, loss and flow length. The full paper is the ICC paper.
- Conference presentationLocation:IEEE International Conference on Communications, Budapest (ICC 2013)Slides:Comments:
- Conference paperPresented at IFIP/Networking
This paper describes the collection and analysis of a very large collection of delay data. The data is collected using the Turbo King method (recursive queries of Domain Name Servers). A model is given to fit the data. The data is analysed by continent.
It is shown that there is a strong relationship between delay and Haversine distance (the distance across the surface of the earth). However, this relationship breaks down in specific cases and occasionally, for some continent pairs, the opposite relationship is shown (longer distance is shorter delay).
- Invited talkLocation:BournemouthSlides:Comments:
This presentation is very like the presentation at Southampton a few months previously. It describes the current incarnation of the Framework for Evolving Topology Analysis.
- Conference paperProceedings of the International Conference on Communications, pp. 2502-2506
This paper tries to establish the relationship between TCP throughput and fundamental qualities like round trip time, flow size and loss. The approach is data driven considering large numbers of publicly available data sources.
- Workshop paperProceedings of IEEE INFOCOM Workshop NetSciCom p.272--277
This paper is the first to describe FETA a process for analysing stochastic models for graph evolution.
- Journal PaperIET Communications, 3(6) p. 957-96
This paper describes the initial results from the MASTS project a measurement project focusing on the JANET UK academic network.
- Workshop paperProceedings from IEEE INFOCOM Workshop, Global Internet Symposium
This paper considers the idea of using fractal space filling curves to locate nearby items in an overlay topology.
- Conference paperProceedings of INFOCOM p. 343-351
- Workshop paperSIMPLEX Workshop at NetSci
This paper describes FETA a method for analysing the likelihood of a particular network topology growing from a given statistical model.