Power BI Shared Datasets versus Cubes

[rthomas_jbg]

Good day Power BI Community.

Question, what do you think about designing, creating and implementing a Shared Dataset as a Cube versus an SSAS Cube ?

I read the article entitled "Power BI Shared Datasets versus Cubes" from the URL link below:

https://www.onesixsolutions.com/blog/data-analytics/shared-datasets-vs-cubes/

I would love to know your thoughts on this.

Thank you.

[Bokazoit]

Hi

Did You move to Power BI datasets instead of using SSAS tabular cube and what was the trade-offs?

I am looking into move to SSAS for several reasons. The PowerBI Pro licens is a shared space between companies making the power to refresh and calculate measures be very controlled by the amount of power available to You at the time. That also means that I am potentially looking into upgrade to a premium version for all users and that is still not enough power vise.

So did You move to datasets or stayed with SSAS?

[rthomas_jbg]

Hello Bokazoit.

Yes I went ahead and created a Power BI dataset modeled with data coming from Microsoft Dynamics 365.

Each user that had access to the data set had a Premium Per user License.

Once you have a properly designed model everything will be fine.

If you think the the Premium Per user License is not enough for you, you can use the Premium Capacity License.

I am still usin the Power BI Dataset.

kind regards,

[GilbertQ]

Hi @rthomas_jbg 

The Power BI Shared dataset is actually also a cube. It is just hosted and mainted by the Power BI team.

If you can get it working within the shared dataset then that is what I would suggest doing.

[rthomas_jbg]

Hi @GilbertQ 

Thank you very much for responsing to my post.

Bsaed on your response to my question, are saying that for example, I can connect to Microsoft Dynamics 365 using Power BI and then design, create and implement a model (Shared Dataset) that can be served as a cube to which the business users can connect ?

[GilbertQ · Accepted solution]

Hi @rthomas_jbg 

yes that would certainly work!

[rthomas_jbg]

Hi @GilbertQ 

Thank you very much for you responses to my questions.
I truly appreciate it.

Kind regards,