Microsoft Dynamics Blog

Welcome to Armanino’s Microsoft Dynamics Blog, where you’ll find information on topics across Microsoft Dynamics 365 ERP and CRM, Dynamics GP, Power BI, PowerApps, Flow and more. Use these tips, tricks and insights to get the most out of your applications! Want these articles delivered straight to your inbox? Subscribe to our newsletter below.

Wednesday, August 8, 2018

Coming Changes to Entity Store and What They Mean for BI Projects

Posted by Noah Kluge

The recent Business Applications Summit in Seattle was another informative Microsoft-led conference.  (There were a lot of different types of sessions, which are now available online for anyone―even those who did not attend.)

A session that I was particularly interested in was presented by Milinda Vitharana and TJ Vassar from the Dynamics 365 for Finance and Operations BI team.  As can always be expected from these guys, there were some good nuggets to take home and reflect on.

A particular area of interest for me is around the Entity Store itself.  In current versions of D365 F&O, this is a high-powered database that underpins embedded Power BI (aka “Analytical Workspaces”) within D365.  The catch is that this database is effectively used only within the context of D365 itself.

The roadmap includes the ability to push this out―think entity export like with Bring Your Own Database (BYODb)―to an environment accessible by the service as well as Azure components (screenshot below).  The new term for this is “Power BI data flows.” In fact, it’s so new the presentation used a prior name!  It sounds like this is a near-term deliverable, perhaps in either October or March.

Image of extending entity store with power bi data flows

What does this mean for your BI project? 

This means that aggregate entities, which are modeled within the D365 development environment, are now accessible outside of D365 itself.  This will enable a cloud-based “Azure only” approach to BI where the data is pushed out to an Azure Data Lake Storage Gen 2 (currently in preview) that can be accessed by Azure Data Factory, Azure ML and other options.

A consideration is that given that the aggregate entity modelling is done within D365, this means it’s a D365 developer task rather than a BI developer task. Often times, these are two different people within a business.

As a Microsoft partner, I look forward to getting my hands on this as soon as possible!


comments powered by Disqus
« | »