Chapter 13. High Availability Cluster Server Information; Reference Information; 7133 Serial Disk Systems; Configuring The Ha Cluster Server System With No Single Points Of Failure - IBM RS/60000 Series Site And Hardware Planning Information

Chapter 13. High Availability Cluster Server Information

This chapter presents information on high availability cluster servers with cluster server cabling
information.

Reference Information

This section has general information about systems or subsystems that can be used in high availability
cluster configurations.

7133 Serial Disk Systems

High Availability Cluster Servers consist of a minimum of two systems in a cluster with two 7133 Serial
Disk Systems. Each system in the cluster comes with AIX operating system software and HACMP high
availability cluster software. The systems use and share the external SSA disks in the 7133 SSA Disk
Storage Systems.
Each system in the cluster has a minimum of two SSA PCI adapters to allow redundant connection to the
7133 SSA Disk Storage Subsystems.
It is highly recommended that a 3153 ASCII terminal (or equivalent) be purchased even if graphics
adapters or remote workstations are used as the control console. This allows a software or service person
to work on one of the cluster servers through the serial port without affecting the other server. If a graphics
display is preferred to the 3153 display, a low-cost workstation can be connected through an Ethernet
connection or a local network.
The 7133 Serial Disk System comes with eight SSA disk drives on two loops and a redundant ac power
supply
Each cluster server can have up to four SSA adapters.
The HA cluster server systems offer configuration flexibility. Because cluster servers are comprised of
systems that can operate independently, all normally available features are supported.
Configuring the HA cluster server System With No Single Points of
Failure
Redundant adapters and mirrored disks are the only way to guarantee redundancy in the 7133 serial disk
system. In this configuration, no single hardware component failure can cause the serial disk system to be
unavailable.
Refer to the High Availability Cluster Multi-Processing for AIX, Version 4.3: Enhanced Scalability and
Administration Guide , order number SC23-4284, and the High Availability Cluster Multi-Processing for
AIX, Version 4.3: Planning Guide , order number SC23-4277, for HACMP/ES planning information.
The following table describes outages and their impacts for the minimum cluster-server configuration with
mirrored SSA adapters (2 Ethernet adapters, 2 SSA adapters, mirrored disk, HACMP, external SSA, and
two power distribution units (PDUs) per I/O rack).
367