Vertipack analyzer and debugging

[UlisesCiccola]

Hello everyone!

I have a question regarding a PBIX file that seems to be really slow when updating a query. I have used vertipack analyzer and get the following. It seems that the table "pedidos" with its column that is primary key ID consumes much of the resources. It is a type text column, and used to match another table also a type text very long char. Any tips on debugging this or also any other tips-resources on how to start reading this vertipack and optimize my model? Thank so much!!!

[v-easonf-msft · Accepted solution]

Hi, @UlisesCiccola 

If you have a needed column  which is the heavier in your  model,the best option is to remove the column and retrieve the same data in other ways. For example, create a composite model keeping that data in a separate table, or create other reports to retrieve that information.

Please check if these related videos are helpful to you :

Reading VertiPaq Analyzer metrics in DAX Studio 

2 ways to reduce your Power BI dataset size and speed up refresh 

Best Regards,
Community Support Team _ Eason

[m3tr01d]

Hi,

just to make sure, when you say 
"I have a question regarding a PBIX file that seems to be really slow when updating a query."

Do you mean when you refreshed the model (update with new date), it takes longer than before OR
Is it when you navigate inside the report (changing slicers, selecting visual), the visuals takes longer to update.

Thank you

[parry2k]

@UlisesCiccola using the whole numbers will 100% change the size. If you have the option for that, please go with it.

[parry2k]

@UlisesCiccola I assume ID is a primary key/unique value. Do you need it in the model? 

[UlisesCiccola]

@parry2k yes, very much needed! Cannot eliminate that column. Would an integer column perform better? But I would have to merge the new column with another table and that could make queries update slower.