Power BI Deployment Pipelines

Analytics is a vital part of decision making, and more than ever creators must collaborate effectively to author reliable and available reports. This session will look at the deployment pipelines tool in Power BI, allowing creators to manage the lifecycle (develop and test) before moving into production.
More on this series can be found at aka.ms/maa

[@monicatanasescu]

Thanks for this video!
Referring to the point you made about having separate datasets and reports, how would you handle a situation, where you have two reports in different stages of the development pipeline?
So for example, you have report A in testing, but you are still developing the dataset for report B. Now, you want to move report B into testing as well, but you don't want the changes to the dataset to affect report A.
Hope that makes sense!
Thanks

[@theacademyhub]

I might have some of the details wrong, but is this accurate?
Report A uses dataset A. You are making changes to the dataset A so it is now dataset B, then make a report B off of dataset B. When you move the B dataset and B report from Dev to Test, you do not want it to affect the original A report and dataset.
If I understand the situation correctly, the only thing that would prevent the dataset from being changed is making a copy of dataset A, call it B as a new dataset. So that way you would have dataset A and report A and dataset B and report B, and then they can be in whatever stages they need to be in the deployment pipelines.
The upshot is, if you make changes to a dataset that affects the report, then the report will need to be updated as it moved from Dev to Test to Prod. Multiple reports off of the same dataset does of course happen often and this is just one of the things to be aware of when doing that.
You can selectively promote items in the deployment pipelines, but you cannot set up multiple pipelines for multiple datasets, otherwise you would have a branching tree of deployment pipelines.

[@__HumanBeing]

Great video, simple and to the point. I see that the demo Dev DB has more records or rows than the Prod DB, is that normal? I thought it had to be the other way around.

[@theacademyhub]

Hi Diego -- you're right, prod systems usually have a lot more data than dev for practical reasons, but in this case, because they were both for examples, the prod just happened to have less than dev. Good eye. :)