[
https://issues.apache.org/jira/browse/HIVE-22122?focusedWorklogId=298574&page=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-298574
]
ASF GitHub Bot logged work on HIVE-22122:
-----------------------------------------
Author: ASF GitHub Bot
Created on: 21/Aug/19 10:26
Start Date: 21/Aug/19 10:26
Worklog Time Spent: 10m
Work Description: XBaith commented on pull request #754: HIVE-22122 Help
improve a little bit of performance and simplified some code
URL: https://github.com/apache/hive/pull/754
I found some areas for improvement by scanning the code.
Maybe contributors could view and merge the code.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
-------------------
Worklog Id: (was: 298574)
Remaining Estimate: 0h
Time Spent: 10m
> TxnHandler.getValidWriteIdsForTable optimization for compacted tables
> ---------------------------------------------------------------------
>
> Key: HIVE-22122
> URL: https://issues.apache.org/jira/browse/HIVE-22122
> Project: Hive
> Issue Type: Bug
> Components: Transactions
> Reporter: Peter Vary
> Priority: Major
> Labels: pull-request-available
> Time Spent: 10m
> Remaining Estimate: 0h
>
> When we do not find open writes for the given validTxnList then we either:
> # do not have any writes on the table - we can return writeIdHwm = 0, and no
> invalid/aborted writes;
> # we have only compacted writes on the table - we can return writeIdHwm =
> nextWriteId -1, and no invalid/aborted;
> # we have compacted writes and some invalid writes on the table - we can
> return the lowest invalid write as a writeIdHwm and set it as invalid.
> What the current code does instead is sending writeIdHwm = nextWriteId -1,
> and sending every write as invalid. This results the same response in case
> 1-2, but probably a longer list in case 3.
> So we have place for some optimizations
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
