vRealize Hyperic server stores monitoring results using a tiered model to minimize the volume of data stored, while still providing sufficient data granularity. Periodically, the vRealize Hyperic server removes detailed metric data from the database and archives it.

. Alerts and events older than a specified age are removed from the database, and not archived.

The server performs the following periodic data management functions.

Compresses and archives measurement data.

vRealize Hyperic server stores detailed metric data (all data points reported) in the vRealize Hyperic database for a configurable period (up to 7 days) of time, after which the metrics are eligible for compression and archival. On a (configurable) periodic basis, the server removes the aged individual metric data points from the database, and archives the metric data in compressed form: hourly metric averages, highs, and lows. vRealize Hyperic server retains the archived metric data for 2 years.

Purges alert data

vRealize Hyperic server retains fired alert data for a configurable period, after which the alerts are deleted.

Purges event data

vRealize Hyperic server retains event data for a configurable period, after which the events are deleted.

Rebuilds metric table indexes

During normal vRealize Hyperic operation, the metric data tables in the vRealize Hyperic database contain a lot of frequently changing data. The vRealize Hyperic server rebuilds the metric table indexes on a (configurable) periodic basis to avoid performance problems that heavily fragmented indexes can cause.