[
https://issues.apache.org/jira/browse/PHOENIX-3590?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15824434#comment-15824434
]
Mujtaba Chohan commented on PHOENIX-3590:
-----------------------------------------
With 5K dense columns and using column names that would be the first and last
columns for immutable encoded and non-encoded scheme so now query time
variation between encoding schemes remains same irrespective of columns
position being filtered on.
Generated 20K rows so data fits in OS page cache when executing query multiple
times. I get the following results:
Query
{noformat}
select /*+ SERIAL NO_CACHE*/ k2,ZZZLAST from $T where A00FIRST='50' limit 1
{noformat}
Average query time for immutable encoded: 2.88 sec and for immutable
non-encoded: 1.93 sec. with is ~50% slower compared to non-encoded.
> Filter on value column for mutable encoded table is > 3X slower compared to
> non encoded table
> ---------------------------------------------------------------------------------------------
>
> Key: PHOENIX-3590
> URL: https://issues.apache.org/jira/browse/PHOENIX-3590
> Project: Phoenix
> Issue Type: Sub-task
> Reporter: Mujtaba Chohan
> Assignee: Samarth Jain
>
> {noformat}
> select /*+ SERIAL NO_CACHE*/ k2,c0,c1000,c2000,c3000 from $T where c4000='50'
> limit 1
> {noformat}
> Query get progressively slower if the column which is filtered is the last
> column of the table.
> For data and schema see data generator script in
> https://issues.apache.org/jira/browse/PHOENIX-3560
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
