what is the most efficient way to update table

[reynold522]

We have a lot of IoT devices uploading data to cloud. 

There several tables

1. Tracking table(overview table): IoT ID, installation data, region, purpose etc.

2. Data table 1: IoT ID, Timestamp, data1, data2 etc. 

3. Data table 2: IoT ID, Timestamp, data1, data2 etc. 

4. DimDate

I linked  IoT ID of tracking table with all data tables and data from DimDate table with all toher data tables. 

My question?

previously, if IT department update one data table/tracking table e.g., add one or more columns, I usually delete old table and load the new table. however, I have to do relink and copy past all measures from old table.

I use directquery in my Power Bi Dashboard. is there any new method to update tables? minimize the effort to update?

Thanks in advance!

[lbendlin · Accepted solution]

Instead of connecting to the tables in your data source, create views and connect to these views.  That way you can shield Power Query from the table structure changes.

[reynold522]

Thank you very much. 

It is well known in snowflake to create view. 

as you talk about create view in power bi, is it via summerize function?

[lbendlin]

Usually you create these views in the Power Query source (the upstream system), likely a SQL Server database .  Power Query queries also act as a type of abstraction/view from the perspective of the Power BI data model.  in DAX you can create calculated tables via SELECTCOLUMNS or similar functions.

[reynold522]

Thank you for help