bitwise.exposed

If you are a decision maker in your company, thinking about if PowerBI is the right decision.

Here’s why, it’s perfectly scalable.

What does perfectly scalable mean? You could start with a zero investment and full performance without restrictions.

What means zero investment? Download PowerBI Desktop from the Microsoft homepage and if you like create also your PowerBI Service account online for free. You’re good to go now.

What does full performance without restrictions mean? Now, our iceberg comes into play.

It doesn’t matter if you are using PowerBI Desktop or PowerBI Service on an enterprise level scale, it’s always the same engine running in the background – Micorsoft Analysis Service. To point it out, as user you see something labled PowerBI on your screen but behind the scene it’s Analysis Service doing the job, that’s the iceberg under the waterline.

Now, what is PowerBI?

There are like two PowerBIs: „Desktop“ and „Service“, but What is the difference between Power BI Desktop and Power BI Service?

Power BI Desktop is primilary for developers. Greatly simplyfied it’s an interface and edtior to an analyics engine called „Analysis Service“ and the ability to display and visualize data. Power BI Desktop is an application, installed on your local workstation, where you could import data, create data models, develop business logic, up to advanced calcualtions (measures) and create user reports. Modelling and developing business logic is a real advantage, especially if you don’t have access to the raw data before import. Or for prototyping, before making adjustments at source level, for example at the data warehouse. Everything you import or develop on PowerBI Desktop, will be saved into a single file (.pbix). It’s one click on the pbix file to open the whole solution in PowerBI, similar like a spreadsheet in Excel. What does that mean? There’s no need for complex infrastructure, setup procedures or similar, Power BI Desktop generates everything in the background and creates a rock solid, single container file. This works for small ad hoc projects, up to enterprise level data warehouse solutions. In principle the hardware is the limit.

Power BI Service is primilary for report consumers. Big difference to PowerBI desktop, it’s not an editor you can’t make changes to the structure or business logic and that’s perfectly fine. This excludes that report consumers couldn’t manipulate or affect the model. Power BI Service is aligned to publish, organize and collaborate.

There are several possibilities, but in a common use case, to publish the data to an audience, you will upload the .pbix-file from Power BI Desktop with one click to the PowerBI Service. Once uploaded it’s accessable via a webbrowser like Google Chrome, Firefox, etc… On the service you don’t have the possibility to make changes to the data model or business logic, but you could create reports and most important you could organize reports in workspaces, group reports together to apps and share them with users. To Power BI Service uploaded reports could be updated and scheduled directly from the source, so you don’t have to upload updates from Power BI Desktop manually.

Performance, volume & pricing

Power BI Desktop

Power BI Desktop – No limitations, it’s your local model, your hardware is the limit and its free. PowerBI provides everything to build and process complex data modells and calculations. Even with an Intel i7, 16 GB RAM modells with up to 500 Mio. rows of data, are processed impressively fast.

Power BI Service

Power BI Free – Equivalent to PowerBI Pro, but it’s free. From my point of view the main difference is the sharing and organizing your data in an enterprise environment, this is not possible with „Free“, but there are some workarounds for sharing. For development, and even small teams it’s absolutly fine to solve a lot of use cases. You could also mix free, pro and premium users, with limitations.

Power BI PRO – Now, it’s no longer free and Microsoft will charge around 15$ per user and month. Model size 1 GB / Refresh 8 per day / max storge 10 GB.
That’s quite a lot and not just a starting point. PowerBI stores data compressed, from 1 GB you could expect 200+ mio rows of data from a complex enterprise level data model. Max storage is the volume of all modells uploaded to the service. Here the same, you could mix free, pro and premium users, with limitations.

Power BI Premium – For „Premium“ price per user is around 30$ per unser and month but a lot more volume then „Pro“. Model size 100 GB / Refresh 48 per day / max storge 100 TB. Due to the much larger possible volumes, „Permium“ is also a littebit more performant than „Pro“, but normally you won’t recongnize that significantly. At least that’s my experience. „Free“ and „Pro“ Users couldn’t access „Premium“ datasets. But normally, it shouldn’t be a problem to restrict the dataset and provide a smaller version to „Free“ and „Pro“ users.

Power BI Premium per capacity – For „Permium per capacity“ it’s around 360$ or 7500$ per month. Model size 400 GB / Refresh 8 per day / „coming soon“. Info about max Storage is currently not provided by Microsoft. In this article details about „Premium per capacity“ is out of scope.

Garnter’s Magic Quadrant for Analytics and Business Intelligence Platforms

There are always pros and cons, therefore here is the link to Gartner, to give you some informations about other BI and analytics platforms on the markt.

Microsoft with PowerBI is leading this challenge since 16 years in a row and Gartner provides pros and cons for each of them.

Make the choice that’s right for you. If you ask me, go for PowerBI and the Microsoft ecosystem. In my opinion, it’s hard to beat PowerBI. The big plus for me is the scalabilty, full performance from day one and extremly fast development, even on enterprise level data models.