Hello, we currently have GoodData.CN 3.21 on AWS and have a question about the Arrow cache with very large datasets. Does GoodData copy data from snowflake to internal arrow storage? Is the data stored in Arrow computed results or raw data? We'd like to compute a result from 500 million rows or more in Snowflake and are wondering about potential system limitations.

Heya Pete, Radek here 🙂 So GD doesn't copy data from Snowflake to the Arrow cache like this, instead we will run the report and cache the results of the SQL query it generates, so it should be much more space-efficient. That said, I can still check for the limitations for you if you'd like!

Thank you, Radek.