Table of Contents
Advertisement
Figure 1-49 Typical cabling scenario for the HMC, the EMS, and the frame
In Figure 1-49, you see the different networks and cabling. Each Frame has two Ethernet
ports on the BPCH to connect the Service Network A and B.
The I/O drawers in which the disks are installed for the EMS Servers also are interconnected.
Therefore, the data is secured with RAID6 and the I/O drawers also are software mirrored.
This means that when one EMS server goes down for any reason, the other EMS server
accesses the data. The EMS servers are redundant from in this scenario, but there is no
automated high-availability process for recovery of a failed EMS server.
All actions to activate the second EMS server must be performed manually. There also is no
plan to automate this process. A cluster continues running without the EMS servers (in case
both servers failed). No node fails because of a server failure or an HMC error. When multiple
problems rise simultaneously, there might be a greater need for more intervention, but often
this intervention does not occur under normal circumstances.
61
Chapter 1. Understanding the IBM Power Systems 775 Cluster
Advertisement
Table of Contents
