Should we use cubes or Power BI models?

[mmossel]

Context

We are now working with on-prem tabular cubes and see the following:

- That making the cubes is labor intensive
- That PBI models are already being made in various places
- That we do not use the extra functionality of cubes (e.g. RLS)

So why keep a central (and somewhat unwieldy) solution when we can achieve more success with decentralized models. My technical heart still cannot accept this (central semantic layer, bundling of knowledge and truth, better management of governance) but I cannot deny that PBI models provide a lot of flexibility and we can use the cubes for the difficult situations (RLS, large datasets). The question is, of course, whether the PBI models / flows are designed for such an enterprise role.

Question

We are thinking to rely more on PBI models and live queries, but we have no experience with this alternative approach. What are the questions we need to ask ourselves to know what is the right choice for us? What Are the Pros and Cons of Cubes vs PBI Models?

Thank you so much

[amitchandak · Accepted solution]

@mmossel , Refer these two options if they can help

https://docs.microsoft.com/en-us/analysis-services/comparing-tabular-and-multidimensional-solutions-...

https://blog.pragmaticworks.com/choosing-to-model-in-power-bi-vs-ssas-tabular

My personal preference is the Tabular model. As Cube can become pretty heavy with the growing needs, as for fast retrieval, they will create a lot of combination

Power BI work Good with Star Schema, Another reason to choose tabular models