The Trickle-Down Effect: Web Caching and Server Request Distribution

Authors:
Ronald P Doyle;Jeffrey S Chase;Syam Gadde;Amin M Vahdat
Affiliations:
Application Integration and Middleware, IBM Research, Triangle Park, USA;Department of Computer Science, Duke University, USA;Department of Computer Science, Duke University, USA;Department of Computer Science, Duke University, USA
Venue:
Computer Communications
Year:
2002

Citing 2
Cited 8

Characteristics of WWW Client-based Traces

Characteristics of WWW Client-based Traces
Web caching and content distribution: a view from the interior

Computer Communications

On the universality of rank distributions of website popularity

Computer Networks: The International Journal of Computer and Telecommunications Networking
Model-based resource provisioning in a web service utility

USITS'03 Proceedings of the 4th conference on USENIX Symposium on Internet Technologies and Systems - Volume 4
Anypoint: extensible transport switching on the edge

USITS'03 Proceedings of the 4th conference on USENIX Symposium on Internet Technologies and Systems - Volume 4
Fingerprint verification on medical image reporting system

Computer Methods and Programs in Biomedicine
On content delivery network implementation

Computer Communications
On using Content Delivery Networks to improve MOG performance

International Journal of Advanced Media and Communication
A quantitative study of Web cache replacement strategies using simulation

Simulation
Modeling content delivery networks and their performance

Computer Communications

Quantified Score

Hi-index	0.24

Visualization

Abstract

Web proxies and Content Delivery Networks (CDNs) are widely used to accelerate Web content delivery and to conserve Internet bandwidth. These caching agents are highly effective for static content, which is an important component of all Web-based services. This paper explores the effect of ubiquitous Web caching on the request patterns seen by other components of an end-to-end content delivery architecture, including Web server clusters and interior caches. In particular, object popularity distributions in the Web tend to be Zipf-like, but caches disproportionately absorb requests for the most popular objects, changing the reference properties of the filtered request stream in fundamental ways. We call this the trickle-down effect. This paper uses trace-driven simulation and synthetic traffic patterns to illustrate the trickle-down effect and to investigate its impact on other components of a content delivery architecture, focusing on the implications for request distribution strategies in server clusters.