[
https://issues.apache.org/jira/browse/CASSANDRA-5043?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13527769#comment-13527769
]
Sylvain Lebresne commented on CASSANDRA-5043:
---------------------------------------------
Using ArrayBackedSortedColumns for collectTimeOrdereData is not correct because
we add sstables results one after the other and so it's not all sorted (see
CASSANDRA-3289). I note that it's possible that adding all the columns from one
sstable to an ArrayBackedSortedColumns and then use addAll (which is allowed)
to add that to a top-level other ArrayBackedSortedColumns might be more
efficient that the TreeMapBackedSortedColumns, but we'd need to benchmark and
I'm not sure it's worth the trouble.
> Small CollationController refactoring and CFS.getRawCachedRow(DecoratedKey)
> fix.
> --------------------------------------------------------------------------------
>
> Key: CASSANDRA-5043
> URL: https://issues.apache.org/jira/browse/CASSANDRA-5043
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Pavel Yaskevich
> Assignee: Pavel Yaskevich
> Priority: Minor
> Fix For: 1.2.0
>
> Attachments: CASSANDRA-5043.patch
>
>
> CollactionController can construct a factory on init based on mutableColumns
> parameter. ArrayBackedSortedColumns proven to be better for counters as well
> as normal columns (row size from 5 to 75 columns). CFS.getRawCachedRow would
> copy key byte array each time it contracts a RowCacheKey so we want to avoid
> that where possible.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
