- Fix regression that could cause queries containing scalar subqueries to fail
- Reduce CPU usage of coordinator in large, heavily loaded clusters.
- Add support for
bitwise_or_agg() aggregation functions.
- Add JMX stats for the scheduler.
query.min-schedule-split-batch-size config flag to set the minimum number of
splits to consider for scheduling per batch.
- Remove support for scheduling multiple tasks in the same stage on a single worker.
node-scheduler.max-pending-splits-per-task. The old name may still be used, but is
deprecated and will be removed in a future version.
- Fail attempts to create tables that are bucketed on non-existent columns.
- Improve error message when trying to query tables that are bucketed on non-existent columns.
- Add support for processing partitions whose schema does not match the table schema.
- Add support for creating external Hive tables using the
external_location table property.