Robust Transform Map for Multiple Classes

[SamiranD]

I have a requirement to load data from a single Import Set Table into three different CMDB classes: Data Center, Computer Room, and Data Center Zone. The target class should be determined based on a field value in the Import Set Table.

Can someone guide me on:

• How to configure a Robust Transform Definition to route records to different CMDB classes based on a field value?
• How does the data flow from the Import Set Table to the target CMDB tables in a Robust Transform?
• Is a separate Staging (STG) table required, or can the Import Set Table be used directly in the Robust Transform Definition?
• What would be the recommended best practice for this scenario?

Any examples or guidance would be greatly appreciated.

Thanks.

[pavani_paluri]

Hi @SamiranD ,

Yes, this can be achieved using a single Robust Transform Definition.

If your source data is landing in one Import Set Table and contains a field that identifies the CI type (for example, Data Center, Computer Room, or Data Center Zone), you can use that field to determine the target CMDB class during the transform process.

In general, the data flow would look like this:

Import Set Table → Robust Transform Definition → IRE → Target CMDB Class

For each record, the Robust Transform evaluates the CI type field and routes the record to the appropriate class:

• Data Center → cmdb_ci_datacenter
• Computer Room → cmdb_ci_computer_room
• Data Center Zone → cmdb_ci_datacenter_zone

A separate staging table is typically not required. The Import Set Table can be used directly as the source for the Robust Transform. A staging table is only useful when additional data cleansing, enrichment, or preprocessing is needed before loading data into CMDB.

From a best-practice perspective, I would recommend:

• Using a single Import Set Table.
• Using a single Robust Transform Definition.
• Determining the target class based on a source field value.
• Loading records through IRE so that identification and reconciliation rules are properly applied.

This approach is easier to maintain and extend in the future. For example, if another facility-related CI class needs to be loaded later, you can simply add the class-routing logic instead of creating additional import processes.

The main thing to ensure is that each target class has the necessary identification rules and required attributes configured so that IRE can correctly identify existing CIs and avoid creating duplicates.

Mark it helpful if this helps you to understand. Accept solution if this give you the answer you're looking for

Kind Regards,

Pavani P