Planning guidelines
Designing the cluster that provides the best solution for your organization requires careful and thoughtful planning. In fact, adequate planning is the key to building a successful PowerHA® SystemMirror® cluster. A well-planned cluster is easier to install, provides higher application availability, performs better, and requires less maintenance than a poorly planned cluster.
You might need to plan for additional processes within your environment. For example, patch management and process management processes are critical if you want your environment to handle various types of failures.
For a critical application to be highly available, none of the associated resources should be a single point of failure. As you design a PowerHA SystemMirror cluster, your goal is to identify and address all potential single points of failure. Questions to ask include:
- What application services are required to be highly available? What is the priority of these services?
- What is the cost of a failure compared to the necessary hardware to eliminate the possibility of this failure?
- What is the maximum number of redundant hardware and software components that PowerHA SystemMirror can support?
- What is the required availability of these services? Do they need to be available 24 hours a day, seven days a week, or is eight hours a day, five days a week sufficient?
- What could happen to disrupt the availability of these services?
- What is the allotted time for replacing a failed resource? What is an acceptable degree of performance degradation while operating after a failure?
- Which failures will by automatically detected as cluster events? Which failures need to have custom code written to detect the failure and trigger a cluster event?
- What is the skill level of the group implementing and maintaining the cluster?
To plan, implement, and maintain a successful PowerHA SystemMirror cluster requires continuing communication among many groups within your organization. Ideally, you should assemble the following representatives (as applicable) to aid in PowerHA SystemMirror planning sessions:
- Network administrator
- System administrator
- Database administrator
- Application programming
- Support personnel
- End users
PowerHA SystemMirror supports a variety of configurations, providing you with a great deal of flexibility. For information about designing for the highest level of availability for your cluster, see the IBM® white paper High Availability Cluster Multiprocessing Best Practices.