Matthias J. Sax created KAFKA-7294:
--------------------------------------
Summary: Optimize repartitioning for merge()
Key: KAFKA-7294
URL: https://issues.apache.org/jira/browse/KAFKA-7294
Project: Kafka
Issue Type: Improvement
Components: streams
Reporter: Matthias J. Sax
For a merge() operator we check at compile time, if one of the input KStreams
requires repartitioning and set the "requiresRepartitioning" flag for the
output KStream for this case. This implies, that we pipe data from all input
KStreams through the repartition topic after the merge().
Using our optimizer, we could push down the repartition operation before the
merge() to only repartition the KStream(s) that required repartition and thus
save network IO for all KStreams that don't require repartitioning.
Note, that this operation is only correct, if all input streams are
co-partitioned (cf. KAFKA-7293).
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
