可扩展机群资源监控系统的设计与实现

论文主要描述了高性能计算机群的资源系统监控系统FARMON的设计与实现。系统通过使用动态加载技术和相应的动态链接库管理协议,使用户可以灵活地添加新的资源监控功能;系统使用LDAP目录服务器,使得系统能对外发布本地机群的资源信息,实现多个机群的协作以及机群和网格计算环境之间的协作;系统使用JavaApplet实现的客户端提供了机群资源的图形监控功能。

Design and Implementation of a Scalable Cluster Monitoring System

This paper presents the design and implementation of FARMON,a resource monitoring system for large scale High Performance Computing cluster.User can freely add new resource monitoring function with its Dynamic Class Load-ing technique and associated dynamic link library management protocol.It uses LDAP directory service server to publish the local cluster resource information to the outside world.So we can implement the corporation of multi-cluster and cluster&Grid computing environment.With the help of Java Applet,it's client provide the visualization of cluster re-source information.