You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Chris Goffinet (JIRA)" <ji...@apache.org> on 2011/02/24 09:09:38 UTC
[jira] Created: (CASSANDRA-2235) Counter AES performance issue
Counter AES performance issue
-----------------------------
Key: CASSANDRA-2235
URL: https://issues.apache.org/jira/browse/CASSANDRA-2235
Project: Cassandra
Issue Type: Bug
Reporter: Chris Goffinet
Assignee: Chris Goffinet
Fix For: 0.8
We noticed tonight when trying out AES for Counters in trunk, there is a serious performance issue when inlining the SSTables. We found that the way we are seeking in the file, BRAF keeps flushing out its buffer of 8MB, and we call dfile.sync() on every row. We are finalizing a patch to write a new SSTable on rebuild, instead of inlining.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] Commented: (CASSANDRA-2235) Counter AES performance issue
Posted by "Jonathan Ellis (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-2235?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12998870#comment-12998870 ]
Jonathan Ellis commented on CASSANDRA-2235:
-------------------------------------------
bq. We found that the way we are seeking in the file, BRAF keeps flushing out its buffer of 8MB, and we call dfile.sync() on every row
fixed in CASSANDRA-2218
> Counter AES performance issue
> -----------------------------
>
> Key: CASSANDRA-2235
> URL: https://issues.apache.org/jira/browse/CASSANDRA-2235
> Project: Cassandra
> Issue Type: Bug
> Reporter: Chris Goffinet
> Assignee: Chris Goffinet
> Fix For: 0.8
>
>
> We noticed tonight when trying out AES for Counters in trunk, there is a serious performance issue when inlining the SSTables. We found that the way we are seeking in the file, BRAF keeps flushing out its buffer of 8MB, and we call dfile.sync() on every row. We are finalizing a patch to write a new SSTable on rebuild, instead of inlining.
--
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] Resolved: (CASSANDRA-2235) Counter AES performance issue
Posted by "Sylvain Lebresne (JIRA)" <ji...@apache.org>.
[ https://issues.apache.org/jira/browse/CASSANDRA-2235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Sylvain Lebresne resolved CASSANDRA-2235.
-----------------------------------------
Resolution: Fixed
Marking resolved since either CASSANDRA-2218 or CASSANDRA-2288 fixed this.
> Counter AES performance issue
> -----------------------------
>
> Key: CASSANDRA-2235
> URL: https://issues.apache.org/jira/browse/CASSANDRA-2235
> Project: Cassandra
> Issue Type: Bug
> Reporter: Chris Goffinet
> Assignee: Chris Goffinet
> Fix For: 0.8
>
>
> We noticed tonight when trying out AES for Counters in trunk, there is a serious performance issue when inlining the SSTables. We found that the way we are seeking in the file, BRAF keeps flushing out its buffer of 8MB, and we call dfile.sync() on every row. We are finalizing a patch to write a new SSTable on rebuild, instead of inlining.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira