cloud cost when using spark bq connector

[kaiseu]

Hi team,

we're running the dataproc job by using the spark-bigquery-connector to interact with bigquery and gcs, while the cloud cost seems a black box to us, do you know how the cost was calculated when using this lib to interact with dataproc/bigquery/gcs? any benchmark results on the cost compared with directly using bigquery sql? and is there any optimization suggestions? thanks