Mailing-List: contact issues-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Date: Fri, 28 Apr 2017 17:17:04 +0000 (UTC)
From: "Eugene Koifman (JIRA)" <jira@apache.org>
To: issues@hive.apache.org
Message-ID: <JIRA.13066879.1493160843000.80635.1493399824136@Atlassian.JIRA>
In-Reply-To: <JIRA.13066879.1493160843000@Atlassian.JIRA>
References: <JIRA.13066879.1493160843000@Atlassian.JIRA> <JIRA.13066879.1493160843737@jira-lw-us.apache.org>
Subject: [jira] [Comment Edited] (HIVE-16534) Add capability to tell aborted
 transactions apart from open transactions in ValidTxnList
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
archived-at: Fri, 28 Apr 2017 17:17:08 -0000


    [ https://issues.apache.org/jira/browse/HIVE-16534?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15989172#comment-15989172 ] 

Eugene Koifman edited comment on HIVE-16534 at 4/28/17 5:16 PM:
----------------------------------------------------------------

you refactored ValidReadTxnList() c'tor and removed the sorting of exceptions - why?
writeToString() always creates 3 ':' - why does the deserializer need cases like _if (values.length < 3) {_

wouldn't be simpler to just serialize the BitSet as "0010110...." - it's very compact and the deserializer wouldn't have to sort and do multiple binary searches.... 

why does _isTxnAborted()_ need a binary search?  why not just look up the in the bitset?

_bitSet.set(0, bitSet.length()); // for ValidCompactorTxnList, everything in exceptio_ - shouldn't this turn all the bits ON?
Nit: seems like ValidCompactorTxnList() c'tor could do this  since it's always the case for compactor


was (Author: ekoifman):
you refactored ValidReadTxnList() c'tor and removed the sorting of exceptions - why?
writeToString() always creates 3 ':' - why does the deserializer need cases like _if (values.length < 3) {_

wouldn't be simpler to just serialize the BitSet as "0010110...." - it's very compact and the deserializer wouldn't have to sort and do multiple binary searches.... 

why does _ isTxnAborted()_ need a binary search?  why not just look up the in the bitset?

_bitSet.set(0, bitSet.length()); // for ValidCompactorTxnList, everything in exceptio_ - shouldn't this turn all the bits ON?
Nit: seems like ValidCompactorTxnList() c'tor could do this  since it's always the case for compactor


> Add capability to tell aborted transactions apart from open transactions in ValidTxnList
> ----------------------------------------------------------------------------------------
>
>                 Key: HIVE-16534
>                 URL: https://issues.apache.org/jira/browse/HIVE-16534
>             Project: Hive
>          Issue Type: Bug
>          Components: Transactions
>            Reporter: Wei Zheng
>            Assignee: Wei Zheng
>         Attachments: HIVE-16534.1.patch, HIVE-16534.2.patch
>
>
> Currently in ValidReadTxnList, open transactions and aborted transactions are stored together in one array. That makes it impossible to extract just aborted transactions or open transactions.
> For ValidCompactorTxnList this is fine, since we only store aborted transactions but no open transactions.


--
This message was sent by Atlassian JIRA
(v6.3.15#6346)