[
https://issues.apache.org/jira/browse/CASSANDRA-11940?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15332921#comment-15332921
]
T Jake Luciani commented on CASSANDRA-11940:
--------------------------------------------
So looks like this is caused by the DeflateCompressor. Since there is no
DirectByteBuffer call for this Compressor it is copying the buffer from mmap to
an on heap then writing the decompressed output to an offheap byte buffer.
Let me prove that's the only issue but in the case of Deflate compressor we
should automatically use non-mmap disk access. Alternatively we can look for a
decompressor library that works with off-heap byte buffers.
> Look into better default file_cache_size for 2.2
> ------------------------------------------------
>
> Key: CASSANDRA-11940
> URL: https://issues.apache.org/jira/browse/CASSANDRA-11940
> Project: Cassandra
> Issue Type: Improvement
> Reporter: T Jake Luciani
> Assignee: T Jake Luciani
> Fix For: 2.2.x
>
> Attachments: abnormal disk read throughput.png
>
>
> CASSANDRA-8464 added support for mmapped decompression where in version <=
> 2.1 the reads were all decompressed in standard heap buffers.
> Since the usage of the file_cache is based solely on the buffer capacity we
> should/can make this much larger in 2.2 when the disk access mode is mmap.
> The downside of this cache being too small is made worse by 8464 since the
> buffers are mmapped/unmapped causing explicit page faults.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
