Converting Data with Integrity: Part 3

Read Part 1 which covers data complexity and intermediate validations. 

Read Part 2 which covers situational requirements and error isolation. 

Conversion Impact 

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.  That is not to say the conversion will never place added load on the source system, but the frequency, duration, and significance is largely determined by the conversion approach.  It is entirely possible for a conversion to be low impact to the organization’s systems and users.  The implementational choices can produce the outcome of a highly disruptive event or one that is not even noticed.  

Establish a Point of Reference 

How to implement a conversion that is transparent to end users?  Avoid continuous interactions with the source system.   Capture content from the source system on a regular interval.  Use the conversion pipeline to establish point in time references. The conversion will operate without obstructing business operations and without competing for system resources.  

 Establishing a point of reference creates the following benefits:

Control

  • Interact with the source systems under controlled and considered constraints
  • Interactions with the source system are reduced to brief moments planned and scheduled to occur at specific times 
  • Selectively interact with only structures holding content relevant for conversion processing 
  • Complete or partial 
  • There are times when a complete snapshot is required, like at the start 
  • At later points, it is possible that only a subset of source system data needs to be obtained 
  • Completes rapidly with most often little to no perceptible impact by on the source system 

Separation 

  • Maintain separation between compute resources dedicated to the conversion and those allocated to service the source system 
  • Creates a point-in-time reference required for effective validations 
  • Simplifies issue identification/resolution 
  • A dated reference facilitates setting expectations about what is contained among source data and identifying content that is not yet included 

The primary goal of implementing a conversion is in maintaining data integrity, but as in life, be considerate of others.  Even when a conversion project completes flawlessly, significant effort is required to ensure there is high confidence in its outcome.  A conversion that causes a disturbance can make that job even more difficult.   

Read Part 1 which covers data complexity and intermediate validations. 

Read Part 2 which covers situational requirements and error isolation. 

For more information about converting data, visit our Data Conversion Services page.

Ready to get started?

Don’t Hesitate to Contact Us

=

Related Tags: