What feature or improvement would you like to see?

When fetching extremely large datasets from snowflake, our current default values for the concurrency level and buffer queue size result in too much data pulled down in the background and causing an out of memory error.

Based on various testing we've done with snowflake, we should update the default values from 10 and 200 to instead be 5 and 100. This will let the defaults work better on most configurations for large datasets before users have to tweak the settings themselves.