摘要
为有效提高集群服务器系统的可靠性、可用性、可扩展性,满足各种应用的需求,采用软硬件结合的设计思想,设计了独立于业务平面的智能管理平台。采用集中管理、分级控制的方法,通过设计两级管理控制器和管理总线,同时定义通用的管理接口协议和数据格式,实现对服务器系统中集群服务器、磁盘阵列、电源等的管理监控和动态配置,为故障判断、定位、隔离和在线维护等提供支持,从而提高资源的利用效率和系统的可靠性。
In order to effectively improve the reliability, availability and scalability of cluster server system and to meet requirements of various applications, by using the design thoughts of combining the hardware and software, an intelligent management platform independent of the service plane is developed, which uses the centralized management and hierarchical control method to define the general purpose management interface protocol and data format at the same time by designing a two-level management controller and the management bus, so as to realize the management monitoring and control and dynamic configuration on cluster server, RAID and power supply in the server system, and support is provided for failure prediction, location, isolation and online maintenance. Thus, the utilization efficiency of resources and system reliability are improved.
出处
《计算机工程与设计》
CSCD
北大核心
2009年第10期2516-2520,共5页
Computer Engineering and Design
基金
"十一五"国防预研基金项目(513160701)
关键词
智能管理技术
服务器系统
智能管理平台
管理控制器
管理总线
intelligent management techniques
server system
intelligent management platform
management controller
management bus
