Normalizing data with the Product Catalog

When data is transferred to BMC Atrium CMDB, data consistency problems like inconsistent product categorization attributes, naming conventions, and duplicate CIs can occur. To ensure that the product titles coming into BMC Atrium CMDB are consistent and aligned, discovered data goes through the Normalization Engine.

As a part of the Normalization process you need to build Normalization aliasing rules to normalize CIs because the discovered data that have uncommon or non standard marketable names. The Categories, Types, and Items (CTI) may need to be updated to reflect the organization's requirement for data classification in the CMBD. These are manuals tasks you need to perform to create and update rules for normalization.

Normalization is performed by checking the Product Catalog, based on matching attributes. If there is no match, the Product Catalog must be updated with any product titles and versions that are not already in the catalog.

Example

Multiple names can be used to represent the Microsoft Office Word product as MS Word, Word 2003, Word, and so on. If you create appropriate aliases, the Normalization Engine can determine that these differences represent the same product. If the approved product name that you have specified in your Product Catalog is Microsoft Word, the Normalization Engine can change these different names to Microsoft Word. If there are different titles for a product are discovered (for example, Microsoft Office 2007, Microsoft 2007.0.01, Microsoft 2007.1), specific normalization rules must be in place to handle them as one product. Because, for example, normalization rules for the market version might not be the same as they are for the manufacturer.

With the Normalization Engine, you can choose which CIs to normalize based on their datasets (such as BMC.ADDM), or classes. When all rules are in place, and all CIs are in sync with the Product Catalog, the normalization process makes sure that product names and categorization are consistent across different datasets and from different data providers.

In BMC Atrium CMDB version 8.0, Version Normalization is always enabled, and it runs before DML Based Instance Normalization. Version Normalization uses the MarketVersion attribute to set a common market version with which licenses are associated. This enables you to store a simpler string version, such as 5.0 or 2007. You can then associate a license with the market version rather than with individual full version numbers.

Also, the Version Rollup rules take precedence over the value in the Product Catalog in setting the value of the MarketVersion attribute in the CI.

For more information about Version Normalization and the normalization process for software license management, see Normalization and software license management Open link .

Info

The complete normalization process involves the following steps:

Configure normalization options — includes set up dataset, class, and system configuration
Configure normalization features
1. Set up Version roll up — allows setting up of rules for populating the MarketVersion attribute
2. Setup suite roll up rules — allows rolling up software discovered as individual components into suites and accordingly consume license per suite rather than per component
Configure Catalog mappings — includes setting up rules for product name normalization and normalized CTI
Create and schedule Normalization jobs to be run at regular intervals using Continuous, Batch or Inline method

When normalization jobs are run, entries can directly be included in the Product Catalog. After normalization is complete and products are in the Product Catalog, you run the reconciliation jobs to complete the process.

The following topics are provided:

Normalizing data with the Product Catalog

Comments