Zoltán Borók-Nagy created IMPALA-14795:
------------------------------------------
Summary: Create union node for partition key scan on Iceberg tables
Key: IMPALA-14795
URL: https://issues.apache.org/jira/browse/IMPALA-14795
Project: IMPALA
Issue Type: Improvement
Reporter: Zoltán Borók-Nagy
When a query is a partition key scan on an Iceberg table, currently we create
the following optimization: for each data file without deletes, only return a
single row.
Instead of creating a SCAN operator and read every data file without deletes,
we could just generate the row batches during planning.
This is something we already do (if enabled) for legacy tables, see
SingleNodePlanner.createOptimizedPartitionUnionNode().
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
