[
https://issues.apache.org/jira/browse/CASSANDRA-5661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13703051#comment-13703051
]
Pavel Yaskevich commented on CASSANDRA-5661:
--------------------------------------------
I think this discussion already outgrown proportions of the original problem, I
want to suggest we try currently implemented queueing approach with expiry and
maximum memory size cap as fix for problem related to memory usage (with LCS in
particular) without introducing any additional complexity on getSegment (e.g.
borrowing) at least for next 1.2 release. And in the meantime I'm open for
discussion of other ways of handling caching which could be borrowing and
gradual control over each instance in the queue, like proposed multiway pool,
without adding any significant overhead on such critical path as reading file
segments.
> 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: CASSANDRA-5661.patch, 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
