Normalization best practices for product catalog implementation

You can choose to normalize data before or after it is written to a dataset in BMC Configuration Management Database (BMC CMDB).

• Normalization batch (scheduled) mode, enabled by default, can be scheduled or started manually from the Normalization Console.
• Inline normalization mode normalizes configuration items (CIs) before they are written to BMC CMDB datasets.
• Continuous normalization mode allows CIs to be written to a dataset before they are normalized.

Recommendations

When you initially load CIs from your data providers into BMC CMDB, BMC recommends that you use the batch mode rather than inline or continuous normalization. Batch mode runs a Normalization Engine job on un-normalized data on demand or on schedule that meets your needs. If you are implementing BMC CMDB for the first time, your current data will not be normalized. After the initial loading of CIs, you can use the inline or continuous mode to update CIs that exist in BMC CMDB but are not normalized.

To make sure that data is normalized initially and kept normalized, use continuous mode to make sure that your dataset remains normalized.

Inline mode is used mainly for integrations when a data source is writing to BMC CMDB, and you might need to take action during the data population process if an error occurs.

For more information about normalization, see Managing consistency of CMDB data by using normalization.