• 中国计算机学会会刊
  • 中国科技核心期刊
  • 中文核心期刊

J4 ›› 2007, Vol. 29 ›› Issue (11): 124-126.

• 论文 • 上一篇    下一篇

网格环境下一种可扩展的资源发现机制

张燕[1] 贾焰[1] 黄晓斌[2]   

  • 出版日期:2007-11-01 发布日期:2010-05-30

  • Online:2007-11-01 Published:2010-05-30

摘要:

在分布、异构的网格环境下,如何快速、准确地查找用户所需要的资源是影响网格计算性能的重要因素。本文针对集中式资源发现方法存在的可扩展性弱的不足,将P2P方法应用到网格资源发现领域,提出了一种扩展性好的分布式资源发现方法。该方法使用二叉树来管理数据,网格中每个节点都负责管理一部分资源信息。实验结果表明,本文提
出的方法能有效解决集中式资源发现方法可扩展性不好、资源信息服务器负载过重、容易造成单点失效等问题。

关键词: 网格计算 资源发现 P2P 二叉树

Abstract:

How to search the resources rapidly and exactly in a distributed, heterogeneous grid environment is the vital dement that affects grid computing perfo  rmance. In this paper, the P2P method is used to deal with the poor scalability of the centralized resource discovery mechanism, and a decentralized res ource discovery method with good scalability is presented. This method uses binary trees to manage data, each node in the grid is responsible for managing a part of the resource information. Experimental results show that the method presented in this paper can resolve many problems that exist in the cen   tralized method, such as poor scalability, heavy load on the resource information server, and single point failures.

Key words: (grid computing, resource discovery, P2P, binary tree)