Slow Performance Through ODBC Connection

Published: 13 Jun 2013
Last Modified Date: 02 Jan 2019


When you have a large data source connected to Tableau through ODBC, you might experience slow performance, particularly while running a large query or creating or refreshing an extract.

In addition, the Processes tab of the Windows Task Manager might indicate that the tabprotosrv.exe process is using an extremely large amount of memory.


  • Tableau Server 
  • ODBC database drivers


In the ODBC Driver Setup dialog box, increase the Prefetch Count value.

This value sets the number of rows at a time that the ODBC driver fetches from the database.


The ODBC driver buffer size is too small.

Additional Information

Some ODBC drivers support a customizable buffer size, for which you can specify the number of rows or amount of memory to allocate for batches of results. The driver uses the buffer to return results to Tableau. Slow performance usually means that the value is too small, and Tableau needs to send frequent queries to the database, increasing the time it takes to populate a view or create an extract.

When you set this value, you want to find the right balance between frequency of queries to the database and amount of memory required for each fetch operation. If you set the value too large, you will experience memory allocation or out of memory errors.

Discuss this article... Feedback Forum
Did this article resolve the issue?