Michael Smith created IMPALA-15313:
--------------------------------------
Summary: Support spilling in the LocalExchanger
Key: IMPALA-15313
URL: https://issues.apache.org/jira/browse/IMPALA-15313
Project: IMPALA
Issue Type: Sub-task
Components: Backend
Reporter: Michael Smith
Some plan shapes - like the probe and build sides of a join or two branches of
a union - require one CTE consumer to get the full result set before another
can start processing it. Other plans may have some CTE consumers that operate
much faster than others. Both scenarios can result in LocalExchanger holding on
to large numbers of RowBatches and consuming memory.
Implement spilling of row batches in the LocalExchanger under memory pressure.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
