[jira] [Comment Edited] (CASSANDRA-5661) Discard pooled readers for cold data

["Jonathan Ellis (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-5661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13694410#comment-13694410 ] 

Jonathan Ellis edited comment on CASSANDRA-5661 at 6/27/13 1:36 AM:
--------------------------------------------------------------------

I don't see what similar-sized files has to do with it.  CRAR buffer size is independent of file size, and that's what causes the fragmentation.  (Zeroing out large buffers isn't free, either.)
                
      was (Author: jbellis):
    I don't see what similar-sized files has to do with it.  CRAR buffer size is independent of file size, and that's what causes the fragmentation.
                  
> Discard pooled readers for cold data
> ------------------------------------
>
>                 Key: CASSANDRA-5661
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-5661
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>    Affects Versions: 1.2.1
>            Reporter: Jonathan Ellis
>            Assignee: Pavel Yaskevich
>             Fix For: 1.2.7
>
>         Attachments: DominatorTree.png, Histogram.png
>
>
> Reader pooling was introduced in CASSANDRA-4942 but pooled RandomAccessReaders are never cleaned up until the SSTableReader is closed.  So memory use is "the worst case simultaneous RAR we had open for this file, forever."
> We should introduce a global limit on how much memory to use for RAR, and evict old ones.

--
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