Forum Discussion
VARCHAR vs NVARCHAR when using DirectQuery in Power BI
Hi Qiuyun,
Thanks for your response. The reason for raising this post is that the client has asked us to improve the performance of a particular table visual in the report. The table visual loads about 100,000 - 200,000 rows depending on a slicer selection. So, I looked at the SQL Server table from where data is being loaded for the visual. The table had couple of NVARCHAR columns and few INTEGER monthly sales columns. I converted both the NVARCHAR columns to VARCHAR columns since they will always store english characters in order to reduce the size of the table and in-turn reduce the size of the data that the report has to read from the database. The NVARCHAR to VARCHAR conversion reduced the size of the table by about 25%(from 480 MB to 360 MB). But after making this change, the performance of the report has come down instead of improving. This is when, I came across the below article which says that Power BI supports only Unicode TEXT datatype. So, Is Power BI trying to convert ASCII strings to Unicode strings which is what is causing the performance degradation in my case ? Is there anyway to avoid this so that I can yield better performance using VARCHAR columns ?
https://powerbi.microsoft.com/en-us/documentation/powerbi-desktop-data-types/
I'm facing same issue. DId you find any solution?