A dynamic server selection method using QoS statistics

DOI Web Site Web Site 4 References

Abstract

<jats:title>Abstract</jats:title><jats:p>In this paper, we study server selection methods from the perspective of selecting the server having the shortest download time when a file is downloaded from a mirror Web site. Specifically, we focus on server selection methods on the client side and propose a dynamic server selection method based on Quality of Service (QoS) statistical data. This method features using past stored data in a more advanced manner than a conventional server selection method by collecting past QoS data according to the day of the week and the time period. We also study and evaluate the method based on experimental results using the Internet. First, we study the optimum values for the parameters of the proposed server selection method. Then we compare and evaluate this server selection method and an existing server selection method. As the evaluation metrics, in addition to objective evaluation indices such as the mean and maximum download time, we use the download give‐up rate and the stress level which reflect the user's feelings. Based on these results, we demonstrate the superiority of this server selection method. © 2004 Wiley Periodicals, Inc. Electron Comm Jpn Pt 1, 87(7): 43–54, 2004; Published online in Wiley InterScience (<jats:ext-link xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="http://www.interscience.wiley.com">www.interscience.wiley.com</jats:ext-link>). DOI 10.1002/ecja.10168</jats:p>

Journal

References(4)*help

See more

Keywords

Details 詳細情報について

Report a problem

Back to top