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
modi123p
Helper III
Helper III

Bug: Data from Power BI Dataset differs when connecting using AS and in Direct Query Mode

Hi,

I tried connecting to a table in Power BI Dataset in Direct Query Mode (local dataset) and compared results by using a Analysis Service connector in import mode.

 

Output from both the queries differ. Output from DQ is fine while the results differ in AS table.

 

Note: This is not an issue regarding difference in refresh time of the import query and DQ output.

 

* I m not applying any transformations except for change type in the query. (All values are showing up in string format in AS query. So applied Change Type to correct the data type)

* Not applying any filters in query/page/visual

* There are no measures in the source table. All fields in the table are selected while importing data. 

*Tried substituting null values if any with 0 for numeric fields.... is this even necessary?

 

Observation:

* Total Sum on numeric fields differ (please refer snapshot below)

* RowCount in AS is different from RowCount in DQ  which indicates missing entries.

 

XMLA_DQ.PNG

 Query Snapshot.PNG

 

1 ACCEPTED SOLUTION

@GilbertQ 

 

Thank you for looking into the issue.

No. The source of data is the same.  Only the connector and connection mode differs.

 

I introduced an index column in the table of original dataset and introduced it as the first column in the import query that used AS connector (to avoid aggregation). 

This resulted in matching values/ entries.

There are null values in some numeric columns which could have caused problem during aggregation.

I tried replacing null with zeros in the original dataset. But for some reason it didn't work!!!

 

I m not sure what is causing the issue.

But the work around of introducing a primary key column as the first column in the AS query seems to be working for me now.

 

Thanks & regards,

Mannu

View solution in original post

3 REPLIES 3
GilbertQ
Super User
Super User

Hi @modi123p 

 

Ok awesome thanks for letting me know.!





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

Proud to be a Super User!







Power BI Blog

GilbertQ
Super User
Super User

Hi @modi123p 

 

From what you have put above there is clearly a difference between the 2 data sources.

 

What happens if you create a report to each one individually, do you get results that match?

 

If they do match then when you are bringing them together something is making the data change?





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

Proud to be a Super User!







Power BI Blog

@GilbertQ 

 

Thank you for looking into the issue.

No. The source of data is the same.  Only the connector and connection mode differs.

 

I introduced an index column in the table of original dataset and introduced it as the first column in the import query that used AS connector (to avoid aggregation). 

This resulted in matching values/ entries.

There are null values in some numeric columns which could have caused problem during aggregation.

I tried replacing null with zeros in the original dataset. But for some reason it didn't work!!!

 

I m not sure what is causing the issue.

But the work around of introducing a primary key column as the first column in the AS query seems to be working for me now.

 

Thanks & regards,

Mannu

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.

Top Solution Authors
Top Kudoed Authors