ODBC connector performance drops when extracting large SQL Server datasets

BA/BI Tableau
, , , , , , , ,
1

We’re experiencing significant performance degradation when extracting large datasets from SQL Server via ODBC connector in Tableau 2023.1. Extract times have increased from 15 minutes to over 2 hours for our main sales database (approximately 8M rows).

The ODBC fetch size seems to be using default settings, and we’re wondering if tuning this parameter could help. We’ve also noticed that network latency spikes during peak hours might be contributing to the slowdown.


ODBC fetch size: 10000 (default)
Network latency: 45-120ms (variable)
Extract duration: 125+ minutes

Has anyone successfully optimized ODBC connector performance for similar scenarios? Should we consider switching to the native SQL Server connector instead? Our reporting dashboards are now experiencing significant delays, affecting business operations.

2

Thanks for the suggestions. We’re constrained to ODBC due to some custom driver requirements for our data warehouse layer. I’ll try adjusting the fetch size parameter. We’re using 64-bit drivers already. The network latency is definitely higher during business hours (9am-5pm), but we need extracts to run during that window for real-time reporting. Any other ODBC-specific optimizations I should consider?