You are viewing a plain text version of this content. The canonical link for it is here.
Posted to reviews@impala.apache.org by "Zoltan Borok-Nagy (Code Review)" <ge...@cloudera.org> on 2026/10/26 15:55:50 UTC
[Impala-ASF-CR] PROTOTYPE IMPALA-15672: Do not truncate the broadcast hash table build factor
Hello Impala Public Jenkins,
I'd like you to reexamine a change. Please visit
http://gerrit.cloudera.org:8080/25086
to look at the new patch set (#2).
Change subject: PROTOTYPE IMPALA-15672: Do not truncate the broadcast hash table build factor
......................................................................
PROTOTYPE IMPALA-15672: Do not truncate the broadcast hash table build factor
With mt_dop > 1, the hash table build cost of a broadcast join is
multiplied by broadcast_to_partition_factor * sqrt(dop), but the factor
is cast to long first. With the default factor 1 and MT_DOP=8,
sqrt(8) = 2.83 becomes 2, so broadcast joins look cheaper than the cost
model intends.
The product is now rounded instead of truncating the factor.
Testing:
- Not measured on its own. No new tests.
Change-Id: I8ea3508f5f6437c0cf3bfef8967e1a9df11b31a4
Assisted-by: Claude Opus 5.5 (Claude Code)
---
M fe/src/main/java/org/apache/impala/planner/DistributedPlanner.java
1 file changed, 3 insertions(+), 1 deletion(-)
git pull ssh://gerrit.cloudera.org:29418/Impala-ASF refs/changes/86/25086/2
--
To view, visit http://gerrit.cloudera.org:8080/25086
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings
Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: newpatchset
Gerrit-Change-Id: I8ea3508f5f6437c0cf3bfef8967e1a9df11b31a4
Gerrit-Change-Number: 25086
Gerrit-PatchSet: 2
Gerrit-Owner: Zoltan Borok-Nagy <bo...@cloudera.com>
Gerrit-Reviewer: Impala Public Jenkins <im...@cloudera.com>