PTC: Proxies that Transcode and Cache in Heterogeneous Web Client Environments

  • Authors:

  • Aameek Singh;Abhishek Trivedi;Krithi Ramamritham;Prashant Shenoy

  • Affiliations:

  • College of Computing, Georgia Institute of Technology, 801 Atlantic Drive, Atlanta, GA 30332 aameek@cc.gatech.edu;Department of Electrical Engineering, Columbia University, 1312 S.W. Mudd, 500 West 120th Street, New York, NY 10027 abhi@ee.columbia.edu;Department of Computer Science and Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400076, India krithi@cse.iitb.ac.in;Department of Computer Science, Computer Science Building, University of Massachusetts, Amherst, MA 01003 shenoy@cs.umass.edu

  • Venue:
  • World Wide Web
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

Advances in computing and communication technologies have resulted in a wide variety of networked mobile devices that access data over the Internet. In this paper, we argue that servers by themselves may not be able to handle this diversity in client characteristics and so intermediaries, such as proxies, should be employed to handle the mismatch between the server-supplied data and the client capabilities. Since existing proxies are primarily designed to handle traditional wired hosts, such proxy architectures will need to be enhanced to handle mobile devices. We propose such an enhanced proxy architecture that is capable of handling the heterogeneity in client needs—specifically the variations in client bandwidth and display capabilities. Our architecture combines transcoding (which is used to match the fidelity of the requested object to client capabilities) and caching (which is used to reduce the latency for accessing popular objects). Proxies that Transcode and Cache, PTCs, intelligently adapt to prevailing system conditions using learning techniques to decide whether to transcode locally or fetch an appropriate version from the server. Our experimental results indicate that the use of PTCs produces significant improvements in the client response times. We show that such results hold true for a variety of data content types like images and video data. Further, we find that even simple learning techniques can lead to significant performance improvements.