Chihiro
Excel Ninja
Read a great article, that articulates why denormalization of db optimized for transactional application is important for business analysis.
https://www.sqlbi.com/blog/marco/2017/10/02/why-data-modeling-is-important-in-powerbi/
I often find that people over complicate their DAX to calculate simple measure from normalized model and impact performance.
For those that use PowerBI for business reporting, I'd imagine you use Data Gateway to securely connect to data source from PowerBI Service. If so, you can install SQL EXPRESS (2012 or later) on the server housing gateway, and use it as staging area for denormalizing data. Then, you can bring data "as is" from SQL EXPRESS to optimize PowerBI performance.
Added benefit of using SQL EXPRESS, is that it supports Direct Query. This will allow for more frequent refresh of data set. As well, you can create Linked Server to MySQL, SQL 2008 or other types of db that does not support Direct Query mode. Thus allowing Direct Query through SQL EXPRESS.
https://www.sqlbi.com/blog/marco/2017/10/02/why-data-modeling-is-important-in-powerbi/
I often find that people over complicate their DAX to calculate simple measure from normalized model and impact performance.
For those that use PowerBI for business reporting, I'd imagine you use Data Gateway to securely connect to data source from PowerBI Service. If so, you can install SQL EXPRESS (2012 or later) on the server housing gateway, and use it as staging area for denormalizing data. Then, you can bring data "as is" from SQL EXPRESS to optimize PowerBI performance.
Added benefit of using SQL EXPRESS, is that it supports Direct Query. This will allow for more frequent refresh of data set. As well, you can create Linked Server to MySQL, SQL 2008 or other types of db that does not support Direct Query mode. Thus allowing Direct Query through SQL EXPRESS.