Populating Axon with Data from other Systems

Challenge

As a quick way to get started using Axon, organizations frequently want to know if they can load data from other systems. For example:

• Systems from a change management database or an enterprise architecture tool.
• Data Sets/Attributes from a data modeling tool like erwin.
• Policies from a policy management tool.
• Glossary items from another data catalog tool.
• Processes from a Business Process Modeling tool

Description

Just because the data can be loaded into Axon doesn’t mean it should be loaded. The Informatica Axon Playbook recommends a “start small, think big” philosophy. Loading hundreds or thousands of unused data assets goes against that philosophy.

• Axon doesn’t replace any tool you have - it isn’t a tool for cataloging all your data.
• Don’t just recreate information already available – For example, IT has lots of tools to help them do their jobs effectively already.
• Axon is there to give the business a way of understanding the data that they see [that’s important to them], recognize and use it, and explain why they need it.

Below are several considerations before loading external data assets into Axon:

1. Only load data that is in scope to be governed. Uploading data assets into Axon that aren’t being governed will result in disconnected, orphaned Axon assets to maintain and navigate until a future phase of the Governance work has them in scope to be scanned and associated to other facet information such as glossary, policies, processes, and so on.
2. Define a process to manage similar asset names. Once the data subjects and data elements are loaded in Axon, if the element name is not an exact match, Axon will create a new entry, unless that element is identified not by name match, but by the Axon facet Reference ID.
   1. The reference ID will need to be used in the case of changing the Attribute name from, for example, “Firstname” to “First_Name” to keep everything already connected to “Firstname” in Axon to the new name “First_Name”.
   2. The Unique ID’s can be exported from Axon (export the search results for Attributes) but someone still must know that “Firstname” was changed to “First_Name” and map the Reference ID to the correct attribute.
3. Develop a process to identify objects deleted from the source system. Once this data is loaded in Axon, how will you know one of them was deleted from the source you obtained it from, if it doesn’t show up in the next upload? Even if a delta check was run, how would a user know the difference between the item being deleted in source system and not being included in the delta update, or the item didn’t have any updates this week/month in source system and therefore it wasn’t included in the delta update?
   
   One can’t assume that just because the item wasn’t included in the delta update that it was deleted.
   
   Axon will continue to have the old attributes unless a process is put in place to notify the Axon data steward/application owner when the data is removed from the source system.
   
   
4. Define the location and procedure to maintain source metadata. Will a procedure be put in place to make either the original source of the data or Axon the source of truth for the data, and only maintain the item’s name, description, and other information in the source of truth system?
   1. Defining this process early on will avoid changes being wiped out in the non-source of truth system when the source of truth system does a subsequent upload with changed information.
   2. How will this process be enforced? Will it be a training solution that Axon users must follow, a technical solution by configuring Axon roles and permissions to limit the end-users who can update this information, or another option?

1. As an alternative, is there an Architecture Review Board (ARB) or some other change management gate that must be passed before systems go into production with the item’s changes?
   1. If so, for the updating of the source system can the items come from those systems and Axon be a check box in that process before a system gets the green light to go into production?
   2. How adamantly is this gate process enforced?

Companies often go through lengthy processes to set up Data Governance tools only to have them fail a year or two later because the data got out date, wasn’t trusted anymore, and therefore was abandoned. That is why it’s so important to ask these questions before loading hundreds of data items from other tools/systems to Axon. They key is in determining how to maintain the data so that it stays current.