08-09-2022 06:12 AM
Most Dataverse implementations, whether they be Power Apps or Dynamics 365 CE on the front-end, contain amazingly valuable data for the end-users and for the organizations' leadership. Although Direct Query is one way to connect Power BI to Dataverse, enterprise deployments will benefit from using Synapse Link feature which is far more appropriate for larger data volumes. Learn the surprising options and best practices (and potential pitfalls) of connecting Power BI with Dataverse though Synapse Server.
It is recommended (but not required) that campers have experience working with Dataverse with either Power Apps or Dynamics 365 CE.
Q: For performance (creating indexes) would it be better to first move the daa to a dedicated pool before querying, serverless queries can be slow
Just to confirm what I answered during the live event, - yes, if you're running the queries interactively and need the performance to respond accordingly, the dedicated pool approach is a step up from serverless in both performance and cost. -
My understanding though (unless I am missing something) is that once I move data to a dedicated pool I have lost the notion of reporting on the near real time data and the dedicated pool is now just a snapshot of my data at a point in time (scheduled refresh)
Was really hoping there was a way to get performance out of the serverless pool queries
That's correct - anything other than Direct Query is going to be a step removed from 'near-real-time.'
Direct Query is the closest to live, but the percieved performance for the report viewer will be significantly slower than a report built off of a cached/import source. - There is a notion of a Hybrid mode in Power BI, that is a cross between import and direct query - it's a jump in complexity that I haven't tested against Dataverse, but it is intended to serve a report with most of the data from a cached source and combine that with direct query of the most recent data.
Is there a best practice to move that data - Polybase?
A documented option is to use the Synapse pipeline Copy exported Dataverse data to dedicated SQL pool - Power Apps | Microsoft Docs -
But there's no reason you couldn't use other supported ETL processes.
Scott Sewell will be keeping an eye on this post for the next week or so. You can drop in your Dataverse questions and keep the conversation going here. 😊
We are getting ready to start! Should be a great show today. Be sure to comment here and we will work to get your questions answered!
The video is not loading.
It's viewable for me here - might have been a temporary miss. -But if you're still unable to view it, the video is hosted on YouTube.
Here's a direct link: Power BI Dev Camp: Bringing Power BI into the Dataverse through the Magic of Synapse - YouTube