Normalization and custom rules

The Normalization Engine provides several out-of-the-box normalization features to enable the normalization of data in your production environment. Because the out-of-the-box normalization features might not meet all your requirements for normalizing the configuration data, you can use the Custom normalization feature to create custom rules that perform set attribute actions based on the conditions that you have defined in a rule. 

The following scenarios illustrates when you might want to create custom rules:

• You have extended the Common Data Model by adding a new attribute. You may want to define custom rules for automatically populating this attribute based on specific conditions.
• You want to make sure that the value of an attribute is consistent across all CI instances in a data set. For example, you can specify that the value of the Name attribute of all virtual systems (where ClassId =BMC_ComputerSystem and isVirtual =Yes) must be stored as uppercase letters.

You can create custom rules for CI and relationship classes. To apply the custom rule to specific instances of the selected class, you can define the qualification criteria. When creating the custom rule, you can also specify whether the rule should be applied to the subclasses of the CI or relationship classes that you have selected. For each custom rule, you must define at least one set of attribute action. You can also define more than one attribute in each custom rule to set multiple attribute values. 

The Custom normalization feature is not enabled for any data set by default. If you enable the Custom normalization feature for an individual data set, the Normalization Engine checks all active custom rules. You cannot select particular rules to apply to individual data sets. 

To use the Custom normalization feature, you must complete the following steps.

1. Define the custom rules for normalizing CIs and relationships. (See Creating a custom rule.)
2. For each data set, enable the Custom normalization feature. (See Configuring normalization settings for datasets.)