About database maintenance

Article:HOWTO64002  |  Created: 2011-12-20  |  Updated: 2011-12-20  |  Article URL http://www.symantec.com/docs/HOWTO64002
Article Type
How To


About database maintenance

You have control over how the data that you gather from monitoring is handled. You can create a custom configuration for the summarization and purging of Monitor Solution data. The data that you choose to store can be as granular or as broad as you require. For example you can specify to store highly detailed and granular data (typically in 5-minute intervals ) for the current day, week, or month. You can specify to then keep less-detailed data that is summarized for the previous days, weeks, or months. Ultimately, the data is purged from the database entirely when it reaches a final age. This control helps you to store the data that you require while also helping you control database growth. However, because data becomes summarized according to a schedule, it also becomes less detailed and therefore less reliable. This factor should be taken into consideration when you schedule the summarization and purging schedule.

Schedule database purging to occur during non-peak times of the day. Large amounts of data can cause the summarization and purging process to take an extended amount of time.

Numeric data passes through the following four stages according to the time-periods that you define:

  • When data is first collected it is stored as fully detailed data for as long as you require.

  • After the specified time period lapses, detailed data is rolled-up into hourly summaries.

  • After the specified time period lapses, hourly summaries are rolled-up into daily summaries.

  • After the specified time period lapses, daily summaries are purged from the database. Database summarization and purging occurs daily at a time you define.

Non-numeric data (such as string metric, process, and NT event data) is not summarized. It is instead stored in full detail for as long as you define and then purged when the specified time period lapses.

The metric polling intervals of a metric can affect the amount of data that is collected. Monitor Plug-in settings and Remote Monitoring Server settings can also affect the amount of data that is stored. For example, you can choose to log NT Event data only when an alert raises. Changing this setting decreases the overall amount of NT Event data that is collected and therefore also decreases the amount of data that is stored.

See Maintaining collected performance data.

Legacy ID


Article URL http://www.symantec.com/docs/HOWTO64002

Terms of use for this information are found in Legal Notices