Published: 22 Nov 2023 | Author: James Beresford

Microsoft's next gen Analytics platform, Fabric went GA last week. With that came the announcement of the pricing structure. It's reasonably straightforward but has a few nuances I'd like to explain.

Capacity based pricing

Fabric's pricing is capacity based, like Power BI Premium. One difference is that there is a Pay As You Go (PAYG) model available for Fabric, where PBI Premium is a fixed monthly cost.

One of the GA Announcements was the Reserved Pricing, which is more cost effective, saving about 41% over PAYG pricing. This brings the Fabric F64 SKU licencing cost in line with Power BI Premium P1 licencing costs, which I had anticipated as otherwise there was a disincentive for existing P1 customers to switch to Fabric.

Fabric Licencing

A key thing to note is that Fabric does have a lower entry price point than Power BI Premium. I'll get into the nuances of Power BI on Fabric slighly later. What this means is that you can start using Fabric at a very low cost. I think this is a lesson learned both from Power BI Premium pricing but also Synapse costs, which were very high from the start, which made adoption difficult.

Estimating appropriate capacity is going to be a bit of an experimental exercise to start with, as there aren't any formal tools available and with the complex combination of workloads it's going to be challenging.

Smoothing

Also worth noting within Fabric is the concept of smoothing - which allows you to consume some of your future allocated capacity if you have a demand spike. This is balanced out by some throttling if you go over.


Power BI licencing

How Power BI licencing intersects with Fabric depends on the Fabric SKU you are on. For those on a Fabric Capacity lower than F64, Power BI Pro licences are still required to view shared content. For F64 and above, it behaves like you have a Power BI Premium SKU in effect - all users in your organisation are licenced to view shared content.

If you are a publisher of Power BI Content, regardless of SKU a Power BI Pro licence is still required to publish, update & share content.


OneLake Storage

Initial indications were that there would be an inclusion of OneLake storage with each Fabric Capacity - that no longer seems to be the case. However as expected, given it is based on ADLSGen2 the pricing is similar (All prices in $AUD per GB):

  • Onelake: $0.0394
  • ADLS Gen2 (Hot): $0.032
  • ADLS Gen2 (Cool): $0.018
  • ADLS Gen2 (Cold): $0.007
  • ADLS Gen2 (Archive): $0.004

What this means from a practical perspective is that in our new architecture for Fabric, we're going to reccomend using Databricks to do ETL, leaving Fabric only for the Gold layer. Placing all your data in an equivalent to hot storage is unneccessary and costly.


Summary

In quick summary:


  • Fabric is purchased by Capacity
  • Storage is separate and priced in line with ADLS Gen 2 Hot tier
  • Power BI Pro licences are still required to publish content

I hope this helps. Keep on top of Fabric News by joing us at the Official Sydney Microsoft Fabric User Group!


Cheers, James

CEO, Talos

Subscribe

Get the latest Talos Newsletter delivered directly to your inbox

TECHNOLOGY PARTNERS

Our partners including Microsoft, UiPath, Databricks & Profisee enable us to deliver business outcomes using best of breed technologies and solutions.

Automation & Analytics Technologies for Business

Our Solutions

CORRIE
CORRIE
Compliance Info Expert

Specialising in all compliance related processes, she has been trained to quickly learn specific compliance processes.

Modern Data Platform
Modern Data Platform
Build Data components

Build, test and implement Data Platform components - secure, efficient, flexible and cost effective.

Automation Initiation
Automation Initiation
Your automation journey

Using our EPIC methodology guiding you to deliver outcomes quickly and cost effectively.

CASSIE
CASSIE
Customer Assistance Expert

Specialising in all customer-related processes, she has been trained to quickly learn specific customer processes.

Click here for more