Automatic Selection of Nearby Web Servers (1998)
| Venue: | in Proc. of the 1998 SIGMETRICS/Performance Workshop on Internet Server Performance |
| Citations: | 4 - 0 self |
BibTeX
@TECHREPORT{Heidemann98automaticselection,
author = {John Heidemann and Vikram Visweswaraiah},
title = {Automatic Selection of Nearby Web Servers},
institution = {in Proc. of the 1998 SIGMETRICS/Performance Workshop on Internet Server Performance},
year = {1998}
}
Bookmark
 |
 |
 |
 |
 |
 |
OpenURL
Abstract
Performance of global services such as the worldwide web can be improved by physically distributed replicas. Most replicated systems today request users to select manually a nearby replica (typically from a list of sites), yet users often are not willing or able to make an informed choice. This paper describes an approach to automatic replica selection which works without change to existing web clients and web and FTP servers. We describe the assumptions behind our approach and propose several metrics for estimation of network distance. We examine the feasibility behind this approach and compare the time required for replica-selection by each algorithm. Finally, we describe a small change to HTTP that would improve replica selection transparency. 1 Introduction A strength of the Internet and the world-wide web is its international character. Web links are location transparent ; users use the same kinds of names to access pages coming from a server in London as from one in New York. A...
