HACMP 5.x 高可用方案培训.pdf

IBM System p

IBM中国技术支持中心
2009-4-10 © 2007 IBM Corporation
IBM System p
目录
– HACMP概念回顾
–常见HA架构
–日常管理
– HACMP 新功能介绍
– Q&A
2 2009-4-10 © 2003 IBM Corporation
IBM System p
Although Hardware is Now Very Reliable, Hardware Failures
Account for a Small Minority of System Outages
ƒ Several studies place the proportion between 20% and 45%
ƒ Human error, software error and planned maintenance cause the
majority of service outages
3 2009-4-10 © 2003 IBM Corporation
IBM System p
HACMP—(High Availability Cluster Multi Processing)
–为什么需要高可用性?
–什么是HACMP?
High Availability:
•系统可用性或运行时间最大化
•系统宕机时间最小化
multi-processing:
•一个cluster里的各个节点上可以运行多个应用
•共享数据或并发访问数据.
– HACMP的目的
•消除单点故障(SPOF),实现高可用
– High Availability is fault resilient
not fault tolerant
4 2009-4-10 © 2003 IBM Corporation
IBM System p
高可用& 容错
Standalone High Availability Clusters Fault puters
Solutions
ƒJournaled File System ƒRedundant Servers ƒLock Step CPUs
ƒDynamic CPU Deallocation ƒworks ƒHardened Operating System
ƒService Processor ƒwork Adapters ƒHot Swap Everything
ƒRedundant Power ƒHeartbeat Monitoring ƒContinuous Restart
Availability benefits
ƒRedundant Cooling ƒFailure Detection
ƒECC Memory ƒFailure Diagnosis
ƒHot Swap Adapters ƒAutomated Fallover
ƒDynamic Kernel ƒAutomated Reintegration
Depends,
Downtime Couple of days In theory, none
but typically 3 mins
Good as your last full
Data Availability Last transaction No loss of Data
backup
Relative Cost 1 2-3 10+
5 2009-4-10 © 2003 IBM Corporation
IBM System p
Software Layers on a HACMP node
ƒ Application
– Uses the services made highly available by
HACMP
ƒ HACMP
– Makes services highly available for applications
– Co-ordinates resource availability through the
cluster
ƒ RSCT
– Provides munication between nodes
– Co-ordination of subsystems
ƒ AIX
– Operating system services
ƒ LVM
– Logical storage management
ƒ TCP/IP