Dataflow Vs Dataset What are the Differences of these two Power BI Components

I have presented about Power BI dataflow and datasets a lot, and always one of the questions I get is: What is the difference between dataflow and dataset. So I thought better to explain it in a post and help everyone in that understanding. In this video, you will learn what the differences between these two components are, when and where you use each of them, and how they work together besides other components of Power BI.
Read my blog post to find links to study more:

*******************

Want to learn about Power BI or AI? or take your Power BI and AI knowledge to the next level for building reports? Check out our courses in RADACAD Academy for all aspects of Power BI and AI.

*******************

********
CONNECT with US!
********

********

[ayandapeter]

I really appreciate this gentlemen's work, I listened to his video on Power Query M vs Dax, that was just perfect explanation. Thanks man

[jeffnixon]

Beautifully done Mr. RR, short and sweet but very much to the point and clear. Thank you for this one and all the other work you do to help we developers deliver power BI solutions!

[wexwexexort]

This was so helpful. Thanks. I have some questions though. I would be happy if you help me clarifiy these:

- From my understanding there are no way to create a dataset using dataflow on the powerbi service. I have to create it using powerbi desktop, then publish the report, right?

- Let's say we don't have complicated ETL or PowerQuery steps. We are handling the EL-T on the data warehouse. PowerBI gets fed by the ready-to-use tables. Do you think it's still beneficial to use the Dataflows?

- I'm also confused by the logic here. We are already doing a data preparation steps while creating the data warehouse. I'm sure everyone does. Then we use refreshs again in the powerbi service. So, it's like doing the same thing twice? Is this services provided by powerbi can be considered like a cloud data warehouse solution? Yet, MS already has Azure Data Warehouse.

- I've been researching about DataWarehouses recently. Can we say this in-memory model powerbi service stores is an OLAP Cube?

[jomzrpg]

Thank you so much for sharing these videos. I have just one question, about performance when multiple reports are using a single dataset as source, does this have impact on performance of the reports connected to that single dataset?

[saratchandra]

how many pbi dataflow connections can we have from desktop at a time ?
we are connecting pbi dataset from get data, where pbi dataset is prepared ?

[danilocampana]

Hello Master,
thanks for generating and sharing quality videos, please
What is the storage capacity of Data Flow?
Would it be the same as Workspace?

[madhukarputta]

Can we use two shared datasets in one report?

[blse]

Very nice explanation, But I have question, Does this mean we can develop an end to end project DWH (Since we are able to do transformation) in power bi? Any insights from any one pls

[feo]

So you can create a dataflow in the Power BI service, but can you also create a dataset in the service as well? I know when you publish a Power BI report from Desktop to the cloud, the visual layer and the dataset goes up to the cloud. But what if you want to create a dataset without a tool like Power BI Desktop?

[ThePradeeppawar]

Hello sir, could you please show one example how to get the data from ServiceNow tool into Power bi desktop

[JJ_TheGreat]

I really do not understand this. Can't I import a file (CSV - which I thought composes a dataset) into Power BI Desktop - and then it connects and then I can ""Transform data" - and then it loads Power Query and I can transform it. YET this is a "dataset", not a "dataflow" (which seem to only be available in Power BI Service). Wouldn't this portion be part of the ETL layer/process, before I load everything into Power BI Desktop by clicking on the "Close & Apply" button on the "Home" tab in Power Query?

Thanks.

[fabiooliveira]

Is it possible to use Dataflows in Powerapps or Powerautomate? Would be really nice to transform data in Dataflow and use it in this solutions.


Another thing, why do we have to do it in live connection? That limitates so much, as example if I want to define my Clients table to be a shared and promoted dataset so everyone use the same client base I wouldnt be able to connect other sourses into it. So if I want to use the same client dataflow in a report that uses tables from sales and in other that uses tables from Marketing that wouldnt be possible.


Is the same thing that happens when I want to use another datasets in usage metrics (connect for example Users Department or plant to see wich department is using it more) I'm not able to because of that restrictions.