Utilizing Connection Usage Characteristics for Faster Web Data Transport

Author:

Moktan Gautam Raj1ORCID,Varis Nutti1,Manner Jukka1ORCID

Affiliation:

1. Comnet, Aalto University, Helsinki, Finland

Abstract

The quest for faster data transport to improve web user experience is ongoing and attempts are conducted from various fronts to realize it. On top of improving user experience, the implications of improving web data transport are also on the energy efficiency of wireless devices as well as user retention rates of service providers. HTTP/1.x allow the opening of multiple TCP connections per server and then using those connections for fetching multiple web objects through the use of HTTP pipelining. With the advent of HTTP/2.0, multiplexing is done inside a single connection to fetch multiple objects. In this paper, we analyze the TCP connections between the browser and the servers and examine their characteristics. We describe how an enhanced TCP variant can take advantage of data transport connection patterns. We show the benefits that enhanced TCP system can bring with the understanding of connection usage patterns. We find that such transport protocol can have effect in the page idle times as well as the connection concurrency during web page transfer. The results show significant improvement of page load times for both encryption heavy and unencrypted pages. We discuss the effect of the transport protocol on object transfer, connection duration, idle times during the page load, connections, and concurrency of flows that cumulate into page load times.

Publisher

Hindawi Limited

Subject

Computer Networks and Communications,Information Systems

Cited by 1 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Corrigendum to “Utilizing Connection Usage Characteristics for Faster Web Data Transport”;Journal of Computer Networks and Communications;2021-02-05

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3