Guest post by Neil Blagrave, Product Manager, BMC Software
First in a series on avoiding the war room
Capacity planners face several challenges:
- Less-experienced staff members have limited knowledge of performance metrics. The workforce is aging, and skilled personnel are retiring or moving to new jobs. Often capacity planners are replaced by individuals who have little or no background in capacity management and who have primary responsibilities elsewhere. The challenge is to find ways to work around this insufficiency of experience, perspective, and time.
- The avalanche of performance data makes it difficult to find and analyze the appropriate data. Ironically, most IT staffs collect and store too much performance data. The key is to collect the right data and to analyze it correctly. Tool vendors typically offer little help in this area – they usually leave it up to you to figure out what data you should be looking for.
- Because data is often collected in technology silos, it has a silo point of view: IMS data, CICS data, DB2 data, z/OS data, and so on. You need a holistic view of the entire IT infrastructure.
By using best practices, you can more accurately predict how changes to IT infrastructure will affect applications and the business in general. You can make better informed decisions, minimize the risk of unplanned outages, and spend more wisely.
Perhaps most importantly, you can avoid appearing in the war room. Like a primary-care physician, you are the generalist who helps the specialists. You are the expert with the integrated, system-wide data that identifies who can solve a problem. With better planning, you can reduce the frequency and duration of war rooms. Both moves will help your career.
Take a holistic approach to capacity management. Look at the performance metrics that matter most. Forecast potential capacity exceptions and when they may occur. Make predictive models showing how metrics such as response times will be affected by changes to the application or infrastructure.
Bridge the technology silos
It’s important to determine how the technology silos are related. A change in one silo may affect the others, and the silos probably share resources.
Consider the simple case of CICS-based applications accessing a DB2 subsystem running in the same LPAR or in the same sysplex. If the CPU utilization of CICS increases, what is the impact on DB2 performance? What is the impact, if any, on the I/O subsystem? What is the impact on the other applications or batch workloads running on the sysplex? A holistic approach ensures that you consider the system inter-relationships when you analyze performance and capacity trends.
In part 2 of this series, we’ll discuss taking an application view of capacity management and how capacity planning affects the business.
The postings in this blog are my own and don't necessarily represent BMC's opinion or position.