Seamlessly Selecting the Best Copy from Internet–Wide Replicated Web Servers

Table of Contents

Seamlessly Selecting the Best Copy from Internet–Wide Replicated Web Servers

A Basic Web Server Model

Benefits and Drawbacks of The Basic Model

Possible Solutions: Caching and Replication

Caching

Cluster Replication

Wide-area (Cluster) Replication

The Technical Challenge

How The Internet DNS works

DNS Round Trip Times Method

Shared IP Address Method

Testing and Results

Evaluation

PPT Slide

Practical Implementation: The Walrus System

The Walrus Architecture

Walrus in Action

The Decision Algorithm in Walrus

Summary

Email: yairamir@cs.jhu.edu

Home Page: http://www.cnds.jhu.edu/projects/webrep/

Other information:
Presentation in the 12th International Symposium on DIStributed Computing (DISC98).