Skip to main content
cancel
Showing results for 
Search instead for 
Did you mean: 

To celebrate FabCon Vienna, we are offering 50% off select exams. Ends October 3rd. Request your discount now.

Reply
DPCCGF
Helper IV
Helper IV

Semantic Model Report Usage in Service

Hello!  I want to do a bit of cleanup on a data model I inherited.  Is there an efficient way to determine what items in a data model are in use by anything in the Service?  If a table or field is not being used would like to somehow identify it and see if I can delete it from the model.  Thanks in advance!

2 ACCEPTED SOLUTIONS
lbendlin
Super User
Super User

How much money do you want to spend on this?  You would have to use Azure Log Analytics against the workspace and track all the queries for some time.  

 

"Not being used"  is a slippery slope. Could also mean "not being used yet" ...

View solution in original post

Jai-Rathinavel
Super User
Super User

Hi @DPCCGF , I will suggest two approaches

1. If you can get a Paid Measure Killer license, then it will be easy to list out the columns, tables or measures not used by any of the report in the data model. If you want to see how measure killer works you can download the free version and test the features on your local model. 

2. Once you enable large semantic model storage setting on your data model. You can use this DMV query ($SYSTEM.DISCOVER_STORAGE_TABLE_COLUMNS) to list out the temperatures of each column in your model. Higher temperature means the column is been queried frequently. You can look out for columns with blank or 0 temperature values and observe the trend for few days and take a call on which columns / tables you want to remove from the model. 

 

Thanks,

Jai 🙂




Did I answer your question? Mark my post as a solution!

Proud to be a Super User!





View solution in original post

2 REPLIES 2
Jai-Rathinavel
Super User
Super User

Hi @DPCCGF , I will suggest two approaches

1. If you can get a Paid Measure Killer license, then it will be easy to list out the columns, tables or measures not used by any of the report in the data model. If you want to see how measure killer works you can download the free version and test the features on your local model. 

2. Once you enable large semantic model storage setting on your data model. You can use this DMV query ($SYSTEM.DISCOVER_STORAGE_TABLE_COLUMNS) to list out the temperatures of each column in your model. Higher temperature means the column is been queried frequently. You can look out for columns with blank or 0 temperature values and observe the trend for few days and take a call on which columns / tables you want to remove from the model. 

 

Thanks,

Jai 🙂




Did I answer your question? Mark my post as a solution!

Proud to be a Super User!





lbendlin
Super User
Super User

How much money do you want to spend on this?  You would have to use Azure Log Analytics against the workspace and track all the queries for some time.  

 

"Not being used"  is a slippery slope. Could also mean "not being used yet" ...

Helpful resources

Announcements
September Power BI Update Carousel

Power BI Monthly Update - September 2025

Check out the September 2025 Power BI update to learn about new features.

August 2025 community update carousel

Fabric Community Update - August 2025

Find out what's new and trending in the Fabric community.

Top Solution Authors