[
https://issues.apache.org/jira/browse/HIVE-26144?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17523145#comment-17523145
]
Janos Kovacs commented on HIVE-26144:
-------------------------------------
For the CTC:
{noformat}
2021-01-12 13:26:57 139947149719296 [Note] The slave is applying a ROW event on
behalf of a DELETE statement on table COMPLETED_TXN_COMPONENTS and is currently
taking a considerable amount of time (61 seconds). This is due to the fact that
it is scanning an index while looking up records to be processed. Consider
adding a primary key (or unique key) to the table to improve performance
{noformat}
HIVE-21917 - COMPLETED_TXN_COMPONENTS table is never cleaned up unless
Compactor runs - was a good help here back-porting, but still saw issues with
full table scan deletes. The index is in production in live-test there and saw
more performance gain than overhead.
> Add keys/indexes to support highly concurrent workload
> ------------------------------------------------------
>
> Key: HIVE-26144
> URL: https://issues.apache.org/jira/browse/HIVE-26144
> Project: Hive
> Issue Type: Sub-task
> Components: Database/Schema
> Reporter: Janos Kovacs
> Assignee: Janos Kovacs
> Priority: Major
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> The following indexes are added to avoid full table-scan in backend rdbms:
> - primary key for COMPLETED_TXN_COMPONENTS
> - primary key for TXN_COMPONENTS
> - index for TXN_WRITE_NOTIFICATION_LOG
--
This message was sent by Atlassian Jira
(v8.20.1#820001)
