Is your feature request related to a problem? Please describe.

Currently, the decision to run a batch query in distributed mode or locally is based solely on the query structure itself, without considering the data size. This approach could pose a problem if the table being scanned is small, as running in distributed mode might incur excessive overhead due to scheduling costs potentially surpassing execution costs. I propose leveraging table statistics (e.g., row count, table size) to estimate an upper bound for IO operations in a batch query. The query should be executed in distributed mode only if the expected IO is sufficiently high.

Describe the solution you'd like

No response

Describe alternatives you've considered

No response

Additional context

No response