Data model filtering

By default, all of the performance metrics and summary and busy hour intervals for a particular technology pack are available to be included in a data model for reporting. When creating a Tivoli Netcool Performance Manager common pack data model, it is recommended to filter these metrics to ensure that only metrics that you need to report on are deployed.

Large data models take longer to generate, merge and deploy and have a larger management overhead than smaller data models. It also requires much more memory to manage a larger data model. The maximum recommended size of a manageable data model is 50 MB, this equates to approximately 40000 metrics. In addition, the larger the data model, the more difficult it is to find metrics in the data model.

Filtering the metrics that are included in each data model by metric group can offer much improved performance, and a targeted set of metrics for particular reporting requirements. If required, you can add additional metric groups by updating the filter and regenerating the data model for inclusion in a new Tivoli Netcool Performance Manager common pack data model.

You can also reduce the model size by filtering out unwanted intervals from your summary and busy hour models.

You work from the CIF file that contains the required raw metrics, busy hours, or summaries from a technology pack or Tivoli Netcool Performance Manager system. Import the CIF file and apply the required filter.