Partial prefetch for faster surfing in composite hypermedia

  • Authors:
  • Javed I. Khan;Qingping Tao

  • Affiliations:
  • Internetworking and Media Communications Research Laboratories, Department of Math & Computer Science, Kent State University;Internetworking and Media Communications Research Laboratories, Department of Math & Computer Science, Kent State University

  • Venue:
  • USITS'01 Proceedings of the 3rd conference on USENIX Symposium on Internet Technologies and Systems - Volume 3
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we present a prefetch technique, which incorporates a scheme similar to data streaming to minimize the response-lag. Unlike previous all or none techniques, we propose partial prefetch where the size of the lead segment is computed optimally so that only a minimum but sufficient amount of data is prefetched and buffered. The remaining segment is fetched if and only when the media is traversed. Thus, it delivers content without any increase in perceived response delay, and at the same time drastically minimizes unnecessary pre-load. The paper presents the scheme in the context of surfing in composite multimedia documents. It presents the technique and optimization scheme used for stream segmentation backed by analytical model and statistical simulation. We report remarkable increase in the responsiveness of web systems by a factor of 2-15 based on the specific situation.