Archives of the TeradataForum

Message Posted: Thu, 19 Jun 2003 @ 18:06:32 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Neil,

A lot depends on how volatile your environment is. Initially, when there are a lot of applications being migrated, you may have to look at your capacity needs several times a year. Later on when you're only looking at (potential) growth of existing applications (both in number of users and amount of data), you can probably go to a yearly capacity planning cycle.

Make sure that you do both (DASD and CPU/IO/Network power) analyses, since they often can grow independently. For example increasing the number of (concurrent) users won't have much impact on space, but will have a dramatic impact on power requirements.

As far as what to analyze... In the case of DASD, you should be looking at object growth, i.e., keep track of current permspace at the table level to see how they're growing. You should be able to extrapolate their size at the end of a year; 2 years, etc. In addition, keep in mind any additional data you expect to accommodate over the planning horizon. In the case of CPU, analyzing AMPUsage data (appropriately expanded to give periodic activity at the user/account level) is probably the most useful. Tallying up the CPU actually used during key activity periods and comparing this to the total CPU available, will give you an idea of how close to the top you are. It can also give you an indication of how much of your workload is currently being delayed because of lack of cpu resource.

Hope this helps.

Pat Belle
Certified Master

	<Prev	Next>		<<First	<Prev	Next>	Last>>