CurtHagenlocher commented on issue #3169: URL: https://github.com/apache/arrow-adbc/issues/3169#issuecomment-3076562575
Broadly speaking, the ADBC driver uses connection and statement options to control the behavior of the gosnowflake library and the Power BI Snowflake connector uses its options to control the ADBC driver. So for a customer to be able to control e.g. the batching or parallelism of fetching the results, these need to be exposed by each layer to the higher layer and then ultimately to the customer. The gosnowflake options are described at [https://pkg.go.dev/github.com/snowflakedb/gosnowflake#section-readme](https://pkg.go.dev/github.com/snowflakedb/gosnowflake#section-readme). It's here that we see that the only way to control the proxy is via an environment variable. This isn't something we can easily manage via the Power BI connector, and to improve on this I think we would need gosnowflake to change to allow direct configuration. (Disclaimer: maybe it does already; I haven't looked at the gosnowflake source.) The ADBC driver options are described at [https://arrow.apache.org/adbc/19/driver/snowflake.html](https://arrow.apache.org/adbc/19/driver/snowflake.html). There is some management of batching available via `adbc.rpc.result_queue_size` and `adbc.snowflake.rpc.prefetch_concurrency` but we're not exposing these in the Power BI connector. That is probably a fairly straightforward change, but the default values also seem pretty generous. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
