Author: Gabriel Russo

Responsibly implementing the conversion is more than moving content from the source to the target.  Conversion processing has the potential to cause significant disruption to system resources. A negative impact to the availability of source data, end-user experience, and system load could be imposed during and because of the conversion.  On the other hand, the conversion could be transparent to end-users without creating an unnecessary burden to the organization.

If intermediate validations built into the conversion pipeline help to ensure the preservation of data integrity, then how does one put together a pipeline? 

With each step of data conversion services, think about how to validate.  If there is an operation that produces a non-deterministic result or if the validation requirements are overly complex, then break up the operation into smaller, more manageable parts.  The conversion should implement intermediate validations.