hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Todd Lipcon (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-4027) Enable direct byte buffers LruBlockCache
Date Thu, 28 Jul 2011 14:22:09 GMT

    [ https://issues.apache.org/jira/browse/HBASE-4027?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13072369#comment-13072369
] 

Todd Lipcon commented on HBASE-4027:
------------------------------------

Nice work and nice write-up. One interesting thing I noticed in the benchmarks is that the
scalability of the FS cache performance is poor - 10 threads looks like it's only twice as
fast as 1 thread on the FS implementations, whereas it scales much better on the various in-process
caches. How large is the dataset in this test? Is it small enough that all of the threads
are contending to read the same HFile block, and thus hitting the locks in HFile reader?

> Enable direct byte buffers LruBlockCache
> ----------------------------------------
>
>                 Key: HBASE-4027
>                 URL: https://issues.apache.org/jira/browse/HBASE-4027
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Jason Rutherglen
>            Assignee: Li Pi
>            Priority: Minor
>         Attachments: 4027-v5.diff, HBase-4027.pdf, slabcachepatch.diff, slabcachepatchv2.diff,
slabcachepatchv3.1.diff, slabcachepatchv3.2.diff, slabcachepatchv3.diff, slabcachepatchv4.5.diff,
slabcachepatchv4.diff
>
>
> Java offers the creation of direct byte buffers which are allocated outside of the heap.
> They need to be manually free'd, which can be accomplished using an documented {{clean}}
method.
> The feature will be optional.  After implementing, we can benchmark for differences in
speed and garbage collection observances.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message