网格资源发现和分配算法研究  

Algorithm Research on Resources Discovery and Resource Allocation

作　　者：刘林东[1,2] 邬依林[1] 

机构地区：[1]广东第二师范学院计算机科学系,广东广州510303 [2]华南理工大学计算机科学与工程学院,广东广州510006

出　　处：《计算机仿真》2012年第12期115-118,250,共5页Computer Simulation

摘　　要：在分布式集群环境中,资源发现和资源分配是网格计算中非常重要的环节。根据丰富的日志记录信息,构造基于日志信息的资源发现和资源分配模型,利用网格环境中各个集群中的守护进程对各子节点的CPU资源使用状态进行分析,提出了一种基于日志的资源发现和资源分配算法LRDA。LRDA算法对比其它的资源发现和资源分配算法能有效地根据日志信息进行资源发现和分配,有效地提高了资源发现和分配的效率,为资源再分配和动态分配提供决策支持。Resource discovery and resource allocation in distributed cluster environment are very important parts in grid computing. According to the detailed log information in the daemon node, a log - based resource discovery and allocation model was constructed, and each cluster daemon can analyse the CPU resources of each child node in the cluster. A resource discovery and allocation algorithm based on Log was proposed. The algorithm based on log information can process resources discovery and resources allocation effectively and it can effectively find the resources in the network compared to other algorithms. LRDA algorithm can provide decision support for the re - allocation of resources and dynamic allocation.

关 键 词：日志 网格计算 资源发现 资源分配 集群 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]