You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Mohamed Mohideen Meeran (JIRA)" <ji...@apache.org> on 2019/01/08 13:06:00 UTC
[jira] [Commented] (HBASE-21632) Backport to branch-1 HBASE-21620
Problem in scan query when using more than one column prefix filter in some
cases.
[ https://issues.apache.org/jira/browse/HBASE-21632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16737103#comment-16737103 ]
Mohamed Mohideen Meeran commented on HBASE-21632:
-------------------------------------------------
[~openinx]
We upgraded hbase from 1.2.5 to 1.4.8 with the patch you provided, in our production servers. We identified a performance degrade in Scanning a long row(around half a million columns) with multiple Column Prefix Filters(around 10 filters). The scan now takes around 1.2 seconds, while the same scan in 1.2.5 took only 0.3 seconds max. We've included the column keys in the attached file(columnkey.txt). Is there a way to increase the performance?
*Table description:*
{NAME => 'MCF', DATA_BLOCK_ENCODING => 'NONE', BLOOMFILTER => 'ROWCOL', COMPRESSION => 'SNAPPY', VERSIONS => '1', TTL => '2592000 SECONDS (30 DAYS)', MIN_VERSIONS => '1', KEEP_DELETED_CELLS => 'TTL', BLOCKSIZE => '65536', IN_MEMORY => 'false', BLOCKCACHE => 'true'}
*Scan query :*
scan 'namespace:tablename', \{ STARTROW => 'row', ENDROW => 'row', FILTER => "ColumnPrefixFilter('1545212621603120001_') OR ColumnPrefixFilter('1546841752667120001_') OR ColumnPrefixFilter('1545387980301120001_') OR ColumnPrefixFilter('1544677866436120001_') OR ColumnPrefixFilter('1546252381017120001_') OR ColumnPrefixFilter('1546247122010120001_') OR ColumnPrefixFilter('2866449000003420001_w') OR ColumnPrefixFilter('1545221612425120001_') OR ColumnPrefixFilter('1546582197395120001_') OR ColumnPrefixFilter('1545798618753120001_')"}
FIle is available in main task(https://issues.apache.org/jira/browse/HBASE-21620)..
> Backport to branch-1 HBASE-21620 Problem in scan query when using more than one column prefix filter in some cases.
> -------------------------------------------------------------------------------------------------------------------
>
> Key: HBASE-21632
> URL: https://issues.apache.org/jira/browse/HBASE-21632
> Project: HBase
> Issue Type: Sub-task
> Reporter: stack
> Assignee: Zheng Hu
> Priority: Major
> Fix For: 1.5.0, 1.4.10
>
> Attachments: HBASE-21632.v1.branch-1.patch, HBASE-21632.v1.patch, HBASE-21632.v1.patch, HBASE-21632.v1.patch
>
>
> Issue for backporting parent.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)