[
https://issues.apache.org/jira/browse/CASSANDRA-7404?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14273870#comment-14273870
]
Ariel Weisberg commented on CASSANDRA-7404:
-------------------------------------------
We are trying to use direct IO for reads to reduce the impact of reads against
the page cache.
For writes we benefit from allowing the page cache and IO scheduler to do their
thing to an extent.
It's not clear yet whether the page cache needs any help in determining hot vs
cold. I am going to circle back to this and see if I can find a benchmark that
benefits.
> Use direct i/o for sequential operations (compaction/streaming)
> ---------------------------------------------------------------
>
> Key: CASSANDRA-7404
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7404
> Project: Cassandra
> Issue Type: Improvement
> Components: Core
> Reporter: Jason Brown
> Assignee: Ariel Weisberg
> Labels: performance
> Fix For: 3.0
>
>
> Investigate using linux's direct i/o for operations where we read
> sequentially through a file (repair and bootstrap streaming, compaction
> reads, and so on). Direct i/o does not go through the kernel page page, so it
> should leave the hot cache pages used for live reads unaffected.
> Note: by using direct i/o, we will probably take a performance hit on reading
> the file we're sequentially scanning through (that is, compactions may get
> slower), but the goal of this ticket is to limit the impact of these
> background tasks on the main read/write functionality. Of course, I'll
> measure any perf hit that is incurred, and see if there's any mechanisms to
> mitigate it.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)