期刊文献+

一种多服务端HTML5 Web Worker迁移系统设计与实现 被引量:7

DESIGN AND IMPLEMENTATION OF A MULTI-SERVER HTML5WEB WORKER OFFLOADING SYSTEM
下载PDF
导出
摘要 计算迁移是解决智能终端资源不足带来的应用性能问题的方法之一。它通过将应用中计算密集型任务迁移到服务端运行来提高应用性能。Web Worker是HTML5标准引入的多线程机制,其面向计算以及相对独立特性使其成为计算迁移重点研究对象。现有Web Worker迁移都是将Web Worker迁移到单种服务端,无法充分利用多种服务端的资源优势。针对Web应用中Web Worker设计一种多服务端Web Worker迁移系统。该系统基于HTML5 Websocket通信机制对标准Web Worker接口进行自定义实现,将Web Worker迁移到单个服务器和由周围终端设备组成的Docker Swarm集群两种服务端。实验结果表明,该Web Worker迁移系统可以充分利用多种服务端资源,有效提升智能终端Web应用性能。 Computation offloading is one of the effective caused by insufficient resources of intelligent terminals. approaches to solve the problem of application performance It improves the application performance by offloading computation-intensive tasks to a powerful server side. Web Worker is a multi-thread mechanism introduced by HTML5 standard. Web Worker is oriented to computing task and relative independent, so it becomes a key research object in computing offloading. All the existing Web Worker offloading methods offload Web Worker to a single kind of server side. It fails to make the best of resources of multi-kind of servers. Hence, we designed a multi-server Web Worker offloading system. By rewriting the standard Web Worker interfaces based on HTML5 WebSocket communication mechanism, we offloaded Web Worker to single server and Docker Swarm cluster composed of smwounding terminal devices. Experiment is conducted and the results show that the implemented multi-server Web Worker offloading system can fully exploit the resources of multi-server side and effectively improve the performance of web application.
作者 王昭 邓浩江 胡琳琳 郭志川 Wang Zhao;Deng Haojiang;Hu Linlin;Guo Zhiehuan(National Network New Media Engineering Research Center,Institute of Acoustics,Chinese Academy of Sciences,Beijing 100190,China;University of Chinese Academy of Scienees,Beijing 100190,China)
出处 《计算机应用与软件》 北大核心 2018年第9期27-31,43,共6页 Computer Applications and Software
基金 国家高技术研究发展计划(2015AA015802)
关键词 HTML5 WEB WORKER Websocket 迁移 多服务端Docker Swarm集群 HTML5 Web Worker WebSocket Offloading Muhi-Server Docker Swami cluster
  • 相关文献

参考文献2

二级参考文献5

共引文献10

同被引文献62

引证文献7

二级引证文献9

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部