Teradata Connector: New (Improved) Connector To Eliminate Duplicate Query to Teradata when Refreshing Native Database SQL Code Based Query
When executing queries with Native Database Queries (with User written / supplied SQL code), Power BI will EXECUTE the same, full length, SQL-based query TWICE against Teradata. This puts extra burden on the Teradata system's resources (CPU, Memory, I/O, etc.) and utilization. When large volume of Power BI reports are deployed and have regularly scheduled refreshes against Teradata, the overhead of performing duplicate queries would cause the company to have to upgrade the Teradata system sooner than if the SQL-based queries were to execute more efficiently.
The request is for a new (improved) Teradata Connector to be developed to address and eliminate the duplicate query behavior.
This is a not a widely known behavior of the current Teradata Connector, because only the DBA or users who have access to Teradata database query logs will be able to see it or notice it.
The same behavior also exists in EXCEL Power Query, if the user chooses to load the data to Data Model also, in addition to loading it to Table (in Excel). No duplicate SQL query is executed if the user only loads the data to Excel Table.
Srinivas Subbarao commented
This is a huge problem at every Teradata customer using PowerBI. I'm surprised that this hasn't gotten traction within MSFT.