作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2008, Vol. 34 ›› Issue (19): 4-5,8. doi: 10.3969/j.issn.1000-3428.2008.19.002

• 博士论文 • 上一篇    下一篇

基于水利信息网络的对等计算平台

王从明,王志坚,王 宇,黄晓萍,程永上   

  1. (河海大学计算机及信息工程学院,南京 210098)
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-10-05 发布日期:2008-10-05

P2P Computing Platform Based on Water Information Network

WANG Cong-ming, WANG Zhi-jian, WANG Yu, HUANG Xiao-ping, CHENG Yong-shang   

  1. (College of Computer and Information Engineering, Hohai University, Nanjing 210098)
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-10-05 Published:2008-10-05

摘要: 我国水利行业对计算力需求不断增加,该文利用P2P技术汇聚PC的空闲CPU资源,来提供低成本的高计算力。针对水利信息网络的现状,以XtremWeb计算平台为基础,构造了基于广域网和局域网的两层XtremWeb网络,并采用多服务窗排队理论进行建模和分析,得到模型的相关参数。给出洪水淹没范围计算的具体实例,采用XtremWeb网络进行实验。实验结果表明,随着Worker数量的增加,实际计算耗时快速下降,系统吞吐量接近线性增长。

关键词: 对等计算, XtremWeb网络, 多窗口排队, 水利信息网络

Abstract: The computing capability has been demanded increasingly in the industry of water resources in China. In this paper, P2P technology is introduced to collect idle CPU resources to provide high computing capability at a low cost. WAN and LAN-based two-layer XtremWeb network is created on XtremWeb computing platform according to the current situation of water information network. Its relevant model is constructed and analyzed with the theory of M/M/N queue to acquire parameters. And the experiments are taken with XtremWeb network on an example of flooded areas. The results show that the actual time of computing declines rapidly and the system’s volume is close to linear growth while the number of worker increases.

Key words: P2P computing, XtremWeb, M/M/N queue, water information network

中图分类号: