Currently, we assign a fixed amount of data to each cloud function worker and use as many cloud function workers as needed for the input data of a query. In this issue, we would support throttling the parallelism of the cloud function workers for a query and exploiting the maximum worker parallelism for a query.