Power Platform Community Forum Thread Details

This might be a dumb question, but I am trying to figure out when to use a data integration vs a connection within PowerApps. Is one a subset of the other? A data integration seems to allow data filtering and manipulation before the data reaches the app. A connection seems to allow triggers and actions using Flow based on what is happening with the data.

However:

- Both seem to allow loading of data to PowerApps

- Both seem to allow saving data from PowerApps back to the source

Can someone clarify and my apologies again if this is a dumb question.

Categories:

Microsoft Dataverse with Power Apps

Assuming I'm understanding your question correctly, a 'connection' is used for ad-hoc queries directly to the data source. So, for example, I can have a SQL 'connection' that lets me directly query SQL data in my Powerapp, etc.

Using Data Integration actually physically copies the data into my CDS environment's 'local' database, so we actually have a copy of the data. So, using the SQL example, we would use the SQL 'connection' in a data integration that brings down some of the data to our app environment. Then our app accesses that local data instead of directly accessing CDS, and the Data Integration takes care of refreshing our data. (although we have to handle updates back to the source database in our powerapp if desired)

Typically we will often use connections in apps UNLESS

- We want to build model-driven apps instead of canvas apps (data needs to be stored in CDS)

- There is some issue (load/performance/etc) associated with 'live' access via a connection.

- There is some reason to operate against a 'copy' of the data instead of directly in production (external processes, etc)