Advertisement

Selecting ISPs

  • Siemiński Andrzej
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6660)

Abstract

The users have now many ISPs (Internet Service Providers) to choose from. Selecting an offer with best price/performance ratio is not an easy task. The key factor is the user perceived latency while displaying the Internet pages. Up to now there is no comprehensive theoretical model of the relationship that links the offered network throughput to the browser latency. Therefore an experiment was necessary to measure the actual browser latency for different ISPs. In the paper several properties of the network connection that is offered by ISPs were recorded and analyzed. The results clearly indicate that some of the popular assumptions do not hold. The network throughput and the browser latency are not tightly bound. The use of Content Delivery Networks does not always result in the shortening of the browser response time. On the other hand the internal (within one IPS) trace routs to a WWW server exhibit substantial stability while they are dissimilar for different ISPs. The analysis of the trace routs suggests that the presence of particular hosts in a trace routes indicates slower than usual download time. This property could be used to select an ISP.

Keywords

browser latency network throughput ISP trace route 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bhatti, N., Bouch, A., Kuchinsky, A.: Integrating User-Perceived Quality into Web Server Design. Computer Networks 33(1-6), 1–16 (2000)CrossRefGoogle Scholar
  2. 2.
    Card, S.A., Moran, T.P., Newell, A.: The psychology of Human-computer Interaction. Lawrence Erlbaum Associates, NJ (1983)Google Scholar
  3. 3.
    Destounis, P., Garofalakis, J., Kappos, P.: Measuring the Mean Web Page Size and Its Compression To Limit Latency and Improve Download Time. J. Internet Research 11(1), 10–17 (2001)CrossRefGoogle Scholar
  4. 4.
  5. 5.
  6. 6.
  7. 7.
    Lamplugh, L., Porter, D.: Charlotte: An Automated Tool for Measuring Internet Response Time. Novell research (1998)Google Scholar
  8. 8.
    Rabinowich, M., Spatschech, O.: Web Caching and Replication. Addison-Wesley, USA (2002)Google Scholar
  9. 9.
    Ramsay, J., Barbesi, J., Preece, I.: Psychological Investigation of Long Retrieval Times an the World Wide Web. Interacting with Computers 10, 1 (1998)CrossRefGoogle Scholar
  10. 10.
    Sevcik, P., Bartlett, J.: Understanding Web Performance, NetForecast, Inc. Report 5055 (2001)Google Scholar
  11. 11.
    Sevcik, P.: NetForecast Formula for Web Response Time Over a WAN, http://www.netforecast.com/
  12. 12.
    Siemiński, A.: Changeability of Web objects - browser perspective. In: Proceedings of 5th International Conference on Intelligent Systems Design and Applications, Wrocław, September 8-10, pp. 476–481. IEEE Computer Society [Press], Los Alamitos (2005)Google Scholar
  13. 13.
    Siemiński, A.: Browser latency impact factors. In: Apolloni, B., Howlett, R.J., Jain, L. (eds.) KES 2007, Part II. LNCS (LNAI), vol. 4693, pp. 263–270. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  14. 14.
    Wang, Z., Cao, P.: Persistent Connection Behavior of PopularBrowsers, http://pages.cs.wisc.edu/~cao/papers/persistent-connection.html
  15. 15.
    Zona Research: The economic impacts of unacceptable Web Site download speeds, White paper, http://www.zonaresearch.com/deliverables/while-papers/wp17/index.html

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Siemiński Andrzej
    • 1
  1. 1.Institute for InformaticsTechnical University of WrocławWrocławPoland

Personalised recommendations