AppDynamics for Databases
2.9.x Documentation
On this page:
The "aggregation interval" is the time in minutes that the AppDynamics for Databases writes out collected performance data to disk, and therefore allow it to be visualized in the user interface. By default, AppDynamics for Databases logs data every minute. Depending on your environment this may be too frequent, and you may be happy with longer intervals.
There are positives and negatives to short aggregation periods, check the list below to help you work out what is best for you.
Positives:
Negatives:
How granular do you need your data?
Check out the screenshot below to compare different intervals.
10 Seconds – 1 Minute – 5 Minutes – 15 Minutes
You can set the aggregation interval either through the UI or if you need more granularity, such as a 10 second interval, you can set the aggregation interval in the dbtuna properties file. Such fine granularity is supported only on the Oracle platform. The aggregation interval is set to 1 minute by default.
Using the collector Setup window, you can set the aggregation window to 1, 2, 5, 10 or 15 minutes.
AppDynamics for Databases supports sub-minute aggregation intervals for the Oracle platform. For example, some customers use 10 second aggregation intervals. To set this use the agg-interval property in the <AppD4DB install directory>/agent/dbtuna.properties file. This is the aggregation interval represented in minutes, however it can be a decimal number.
For example for
Note: A sub minute aggregation interval incurs additional overhead both on the AppDynamics for Databases server and the Oracle database. Test to ensure that your systems can handle the additional overhead and upgrade capacity as required.
This is the section of the dbtuna.properties file that controls the sampling and aggregation intervals.
# Collector Settings agg-interval=.1667 sample-rate=999 samples-per-second=1 cpu-miliseconds=602