跳到主要导航 跳到搜索 跳到主要内容

Evaluation of data streaming effect on real-time content transformation in active proxies

  • National University of Singapore

科研成果: 书/报告/会议事项章节章节同行评审

摘要

In this paper, we would like to quantify the effects of different modes of real-time content transformation in web intermediaries based on the ultimate performance matrix - the client's page retrieval time. The possible three modes of study are byte streaming, whole-object buffering, and chunk streaming. Due to the potential "pushing back" effect to the retrieval of embedded objects in a page, it is found that their performance might vary substantially. While byte streaming transformation might not be practical, the whole-object buffering strategy can easily cause an average increase of 10.45% and 41.67% in page and object retrieval latency respectively, independent of the exact transformation algorithm involved. This study not only gives the fundamentals for real-time content transformation in web intermediaries, but it also serves as an important implementation guideline for value-added network services.

源语言英语
主期刊名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
编辑Laurence T. Yang, Minyi Guo, Guang R. Gao, Niraj K. Jha
出版商Springer Verlag
1055-1064
页数10
ISBN(印刷版)354022906X, 9783540229063
DOI
出版状态已出版 - 2004
已对外发布

出版系列

姓名Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
3207
ISSN(印刷版)0302-9743
ISSN(电子版)1611-3349

学术指纹

探究 'Evaluation of data streaming effect on real-time content transformation in active proxies' 的科研主题。它们共同构成独一无二的指纹。

引用此