Configuration Rules for a Mixed HP 9000 / Integrity Serviceguard Cluster, March 2007

Table Of Contents

Figure 7. New four-node mixed HP 9000 / Integrity cluster

Adding one critical application and two new Integrity nodes to existing

four-node HP 9000 cluster

This example shows a homogeneous four-node HP 9000 cluster with one mission-critical application

on each node. Because of resource constraints and new application deployment, the cluster must be

upgraded.

Original four-node HP 9000 cluster with four critical applications

Each node is configured with one mission-critical application:

• Human Resource (HR)

• Enterprise Resource Planning (ERP)

• Financial (FI)

• Network File Sharing (NFS) system

All applications, except NFS, have their own database. For the purpose of this example, it is

assumed, that the databases deployed in the cluster don’t support failing over among heterogeneous

cluster nodes. Hence the NFS server is the only package that is allowed to fail-over among

heterogeneous nodes. The cluster is configured symmetrically, which, in theory, allows any host to run

any application. In reality, each node can hardly meet its performance expectation by just running

one application. Reduced performance must be accepted in a failover scenario when just two

applications are running on one node.

Node B

as a

failover

database

server

Database Server 1

App Server *

Database Server 2

App Server *

Failover

NOT

IPF

Node A

Node B

Node C

Node D

Failover