Is it better to create a view in the database and have Power BI connect to that view, or would it be more viable to create parameters to filter data/queries?
Say I have a table which is fairly massive but I want to report on it per 'region' (based on a dimension in table). Currently, I create a view in the database and then connect Power BI and refresh the charts and it seems to work well. As more 'regions' become interested in having their own dashboard, I have to create and manage additional views though. It would be nice if the end-user could filter for their own region and have the data refresh.
- The table with all 'regions' accumulates many rows per day and is >5mil rows total so I would want the view to load quickly
- The access to the data is through PowerBI.com - I publish the dashboard there so the end user would have to be able to select a parameter through the web portal. No one has access to the underlying database.
- I assume using a Power BI template (empty dashboard with no data) and parameters would not work unless my end users had direct access to the database, right? So if I created an Azure SQL Server database and pointed Power BI to that source, an end user could have Power BI visualize their 'queries' (through pre-built parameters)
