Power BI – How Power BI Integrates into Fabric

By: David Rohlfs

Introduction

There have been a lot of changes recently to the Microsoft Power Stack. Microsoft has been releasing their new SaaS (Software as a Service) called Microsoft Fabric. Fabric is meant to be the solution to data and data storage.

Since the release of Microsoft Fabric, there have been a lot of changes to how Power BI is structured and especially the Service side of Power BI. This blog aims to explain some of these changes and focus on how current Power BI users are and will be affected by the release of Microsoft Fabric.

Relevant Disclaimer:

I am a Power BI Report Developer and am considered an expert at Power BI. I am NOT an expert data scientist or data engineer. My focus with Fabric is to be able to migrate and utilize the Fabric SaaS if I need to. If you get into reading about the specifics for Fabric, it is really cool, but if your only focus is Power BI, there is a lot of stuff that you just don’t need to worry about.

Power BI and Fabric

Let’s start off by explaining what Microsoft Fabric is. Fabric is the interface that connects users to the Microsoft One Lake and is meant to contain the data science, data engineering, data analytics, and business intelligence needs all in one SaaS. In basic terms, Fabric is meant to be a one-stop-shop for all data needs.

Over about the past year Power BI has been intertwined with Fabric. There are some obvious changes like web URL’s and the word Fabric popping up in new spots, but there are also some changes that you may not have noticed. The biggest of these changes is that Power BI has been migrated into Fabric space already. What I mean is that Power BI can already connect to Fabric through workspaces, semantic models, reports, dataflows, data warehouses, KQL stuff, and Fabric notebooks. If you are familiar with the Power BI service, then the Fabric service will be similar and not too hard to navigate. Microsoft has set up Fabric to use Workspaces, Tenants, Domains, Dataflows, and Semantic Models. These are things that experienced and inexperienced Power BI users have come across, even if you don’t know it.

Let’s focus on workspaces in Fabric. If you have a Power BI Pro account, you have probably created a workspace before. The process is essentially the same in Fabric and has similar components to it. In the Power BI service, your focus in a workspace would be to place reports, semantic models, dashboards, and dataflows (along with a couple of other less common items). The Fabric workspace does all of these things along with other data engineering and data science items like notebooks.

So, what about the data? This is where Fabric starts to get amazing with Power BI. The focus of Microsoft Fabric from the data storage perspective is to have everything ‘under one roof’. Meaning that you can keep all of your data storage inside of the One Lake and you don’t need to copy data from one data source to another. You can also create “Shortcuts” in Fabric for the outside data sources that you don’t want to migrate to Fabric.

Note: I am not going to explain what a shortcut is in this blog, but it is worth learning about if you are a Power BI report developer. I expect in the near future of Power BI, shortcuts are going to be a very common thing that Power BI report developers are in charge of.

Time to talk about the migration of the Power BI Admin to being the Fabric Admin.

Microsoft has already moved the admin portal from Power BI to Fabric. The portal is similar to how it was while it was still in Power BI, but as Fabric keeps being developed the portal will likely undergo more changes. If you were a Power BI admin for your tenant, then you have automatically been migrated to being the Fabric admin for your tenant.

Common Problems

The Fabric Push Towards PySpark

If you have been a part of any Fabric training, you probably noticed the intense focus on PySpark and Notebooks. From a Power BI perspective, I don’t really care about learning PySpark. This isn’t because it isn’t powerful, but because I have learned and will continue to do my transformations inside of Power Query. While PySpark would be good to learn for a Power BI developer, I don’t see it as something I need to learn right away or in the near future.

Constant Updates from Microsoft

If you are reading this blog, it is probably already out of date. Microsoft has been changing and updating Power BI and Fabric a lot lately. There are so many updates just to Power BI that it can be hard to keep track of what to follow and where to go. If you are someone who doesn’t like to have all of the preview features in Power BI, you should know that Power BI is forcing users to have these changes on their desktop app. If you don’t know about Preview Features, I will link a blog below. Anyways, the changes to Microsoft are becoming so common that it can be hard to keep up with everything that is happening.

Fabric Pricing

One of my biggest complaints with Fabric is that the price constantly seems to be changing. Because Fabric is still brand new and is going through so many constant updates, there isn’t a great way to know exactly what price you will need to pay to be in Fabric. What I can say is that a lot of the general pricing has become stabilized. So, if you want a rough estimate of what it would cost to migrate your organization to Fabric, you can find that amount. But, because Fabric is so new, it might be hard deciding which license to obtain this early on.

Conclusion

My hope with this blog was to keep it strictly concerning Power BI. A lot of the documentation around Fabric has been focused on the perspective of a data scientist or data engineer because these are professions where Fabric is creating a huge impact. Thankfully for Power BI, Fabric has made some major changes, but it has not redone the general process. There are a lot of topics concerning Power BI and Fabric that I have not covered in this blog. I highly recommend spending some time reading or watching some Microsoft explanations of Fabric.

Links Related to This Blog:

Power BI Preview Features

Comments