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

Earn the coveted Fabric Analytics Engineer certification. 100% off your exam for a limited time only!

Reply
brownrobm
Frequent Visitor

Memory Allocation Failure: Large number of relationships

I have been using Power BI for a while and I have never run into something that it cannot do. I have loaded large amounts of data into it and it has handled it without any problems. However, I am getting the "Memory Error: Memory Allocation Error.Try simplifying or reducting the number of queries." This is on a 64-bit Windows 10 desktop with 8 GB of ram. The memory usage on the box is 4GB while this is running.

 

There seems to have an interernal limit on the number of relationships that it can handle.

 

I have 2 tables: Association and Organization.  Organization has about 70k organizations in it and the table is fairly wide with 20 or so columns. Association has about 140k relationships between those businesses and is made up of ParentOrgID, ParentName, and ChildOrgID. The entire model is well under 100 MB. I am just trying to create a relationship between Association.ChildOrgID and Organization.OrgID. That is when the memory error occurs. An organization can belong to many different associations and a single Association can have 8k Organizations.

 

Any ideas on how to work around this issue? I want to be able to pick an association and see all of the organizations that belong to it.

 

Any help would be appreciated! Thanks!

 

 

 

1 ACCEPTED SOLUTION

Finally found a solution to this. 

 

I changed the query that was pulling my associations down to only pull a few thousand rows. I then created the relationship (with no memory error). Finally I changed the query back to the original form that pulled down the full dataset (139k rows). PowerBI handled it without any problems.

View solution in original post

6 REPLIES 6
v-yulgu-msft
Employee
Employee

Hi @brownrobm,

 

Did you use the 64-bit version of Power BI desktop?

 

If the Query Editor query is overly complex, such an error would occur. Please remove those unnecessary columns and delete the problematic step in Query Editor. Then, try again.

 

Regards,

Yuliana Gu

Community Support Team _ Yuliana Gu
If this post helps, then please consider Accept it as the solution to help the other members find it more quickly.

Yes, I am using the 64-bit version of Power BI Desktop. The query editor is not overly complex; there are only 2 SQL Select statements being run. Those SQL statements are large, but they execute on the SQL Server, not on my desktop and I would not think they would affect the performance of my Power BI Desktop.

 

Again, I am not having the problem in the query editor. The problem occurs when I attempt to create a relationship in the model.

Finally found a solution to this. 

 

I changed the query that was pulling my associations down to only pull a few thousand rows. I then created the relationship (with no memory error). Finally I changed the query back to the original form that pulled down the full dataset (139k rows). PowerBI handled it without any problems.

Anonymous
Not applicable

Hi. I am having same problem. 28 tablaes, some with more than 20 columns and at leats 300k row. 

So, should i have limit the number of columns/ row? .. and how do i change that since i have already exported the data in the power bi model? 

 

Looking forward.  Thanks.

I would change my exports to 1k rows per table. That is similar to what I did in order to make mine work.

Anonymous
Not applicable

So.... would it still be possible even after exporting the objects to PoweBi Modelo?

How in query can do do that since my tables have been imported it. Sorry the ignorance, freshly new in PowerBi. 

Helpful resources

Announcements
April AMA free

Microsoft Fabric AMA Livestream

Join us Tuesday, April 09, 9:00 – 10:00 AM PST for a live, expert-led Q&A session on all things Microsoft Fabric!

March Fabric Community Update

Fabric Community Update - March 2024

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