neilconway opened a new issue, #23982: URL: https://github.com/apache/datafusion/issues/23982
### Is your feature request related to a problem or challenge? For every batch, `BoundedWindowAggExec` in `Linear` mode does two operations that are `O(n)` in the number of live partitions: - `aggregate_evaluate_stateful` / `evaluate_stateful` iterate all of `PartitionBatches` and probe `PartitionWindowAggStates` for each partition, per window expression - `update_partition_batch` calls `set_most_recent_row` on every buffered partition This is pretty inefficient when the # of partitions significantly exceeds the batch size; most partitions won't receive a row in the current batch, so we shouldn't need to touch them. ### Describe the solution you'd like _No response_ ### Describe alternatives you've considered _No response_ ### Additional context _No response_ -- 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: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
