SYNCHRONIZING THE HUB AND SOURCE SYSTEM
The initial population of the CDI hub requires loading detail from each of the source systems. This activity is typically performed as a bulk data transfer from each of the source data systems.
The actual logistics of the hub loading typically requires the construction of flat file extracts from each of the operational systems that in turn are loaded into the CDI hub. Exhibit 7.7 shows the loading of each data source in succession: (1) service, (2) billing, (3) sales order, and (4) loyalty. The CDI hub will integrate the individual customer records as it loads each of the sources.
For a registry-style hub, only those data elements required for customer identification are loaded into the hub. For persistent hubs, both the identification and the descriptive elements are loaded into the hub. The actual time required to perform the initial load and integration may span several hours (or even days, depending on the data volumes).
Once loaded, the CDI hub’s data can be maintained or synchronized with source systems in a transactional manner. This typically requires custom development in order to capture any source system changes and ensure they are propagated to the CDI hub. (This is where leveraging an existing EAI infrastructure can simplify CDI hub implementation). As data is updated on a source system, a transaction can be applied against the CDI hub to update the customer’s detail.
EXHIBIT 7.7 Loading Four Sources into the CDI Hub ...
Get Customer Data Integration: Reaching a Single Version of the Truth now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
