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

July 7 - July 17 | Round 2 of the Power BI Dataviz World Championships. Don't miss your chance! Learn more

Reply
a4apple
Helper I
Helper I

Data Modeling Question

Hello Everyone,

 

I am working with Orders and OrderItems model. There are 6 different individual dimension keys that make up a primary key including a date field. To effectively use the filters I am using the linking table concept (Created a Primary key in the linking table via ETL and using that table to filter the facts). 

 

Basically my model is like,

 

Orders <==> OrdersDimension <==> Rest of Dimensions

OrderItems <==> OrdersDimension <==> Rest of Dimensions.

/t5/image/serverpage/image-id/3095i6752181269DACC47/image-size/large?v=1.0&px=600

Works fine, but with historical data, I have more than 2Billion unique PK records via ETL which is a limitation in Tabular.

 

How can I get around this limitation?

 

I have tried another solution like linking the dimensions to the fact tables directly. but I have to write complex DAX to propagate filter conditions to OrderItems measures.

 

https://stackoverflow.com/questions/10181316/2-billion-distinct-values-per-column-limitation

 

Basically to solve a problem like this

https://community.powerbi.com/t5/Desktop/Best-Way-for-work-with-Multiple-Fact-Tables/td-p/21441


I saw an answer provided in this solution but I have no clue how to merge?

 

Any suggestions are appreciated. Thanks in advance.

 

 

2 REPLIES 2
GilbertQ
Super User
Super User

I would suggest looking to aggregate your data, as often when I have been working people do not go down to the very granular data.

This will ensure that your model is smaller, as well as possibly looking to merge some of the data into smaller tables?




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

Proud to be a Super User!







Power BI Blog

Thank you @GilbertQ . At this moment that is one of the options I am considering. I am also trying to understand if there is any other work around for this situation meanwhile.

Helpful resources

Announcements
FabCon and SQLCon Barcelona 2026

FabCon & SQLCon – Barcelona 2026

Join us in Barcelona for FabCon and SQLCon, the Fabric, Power BI, SQL, and AI community event. Save €200 with code FABCMTY200.

60 days of Data Days Carousel

Data Days 2026

Join Data Days 2026: 60 days of free live/on-demand sessions, challenges, study groups, and certification opportunities.

Power BI DataViz World Championships carousel

Power BI DataViz World Championships - June 2026

A new Power BI DataViz World Championship is coming this June! Don't miss out on submitting your entry.

Top Solution Authors