Data Flows
Data Flows define the basic requirements of an Integration Project. It's one thing to say "integrate X to Y". Data Flows put some high level definition around that by expressing what types of data will flow from which system to the other. Ultimately, this determines the size of the Project.
Note: All Data Flows are described in one direction. If you want to bidirectionally sync any data with your integration, define two separate Data Flows with opposite Source and Target Endpoints.
Configuration Options
General
- Data Flow Name - Name the Data Flow something identifiable throughout the application. A good template to use is
<Data Type> from <Source> to <Target>which succinctly expresses what it does. At the very least, you'll want to describe the entity or entities being integrated in business terminology. - Source Endpoint - Which application is the integrated data starting from? (Either the Source or Target should be an "Owned" Endpoint.)
- Target Endpoint - Which application is the integrated data moving to? (Either the Source or Target should be an "Owned" Endpoint.)
- Expected Flow Complexity - How complicated do you think it will be to move that data? Is it complex, highly configurable data or very straightforward simple data? Are data formats different (e.g. XML on one side and JSON on the other)? Are there caluclations involved? This field takes t-shirt size estimates, so it's intended to be imperfect. It's more important that you apply those t-shirt sizes consistently. If you completely don't know, we selecting
Medium. - Extpected User Input Complexity - How much user input will be required or enabled for moving this? Will every one of your users get the same integration or will they have the opportunity to customize it? Do you think they'll be required to customize it?
How This Is Used
Data Flow information is used to establish the size of a Project. Each Data Flow calclates an Effort Score. This is similar to an Agile story point, but weighted by the factors you enter and the Endpoints involved.
While Data Flow Effort Scores are meaningless on their own, a Data Flow that has an Effort Score of 8 should be considered "four times as big" as one that is scored a 2. This same logic applies when comparing Projects, which are scored by summing the Effort Scores of each Project's Data Flows.