[
https://issues.apache.org/jira/browse/HADOOP-11867?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17264930#comment-17264930
]
Steve Loughran commented on HADOOP-11867:
-----------------------------------------
Owen, I'm doing some work on openFile (HADOOP-16202) and I'm thinking about
vector IO there
* the read policy there (random, sequential) should a "vectored" option to say
"plan for vectored IO"
* your PR should add a new mandatory option for FileOpen which can be used to
declare that vectored IO must be supported by the stream, e.g
must("fs.option.openfile.vectored", true). Any FS which doesn't recognise
it/support it will fail fast in openFile, rather than wait until the IO. I'm
thinking of things like ftp which don't support seek(). Everything else should
work, shouldn't it?
> FS API: Add a high-performance vectored Read to FSDataInputStream API
> ---------------------------------------------------------------------
>
> Key: HADOOP-11867
> URL: https://issues.apache.org/jira/browse/HADOOP-11867
> Project: Hadoop Common
> Issue Type: New Feature
> Components: fs, fs/azure, fs/s3, hdfs-client
> Affects Versions: 3.0.0
> Reporter: Gopal Vijayaraghavan
> Assignee: Owen O'Malley
> Priority: Major
> Labels: performance, pull-request-available
> Time Spent: 6.5h
> Remaining Estimate: 0h
>
> The most significant way to read from a filesystem in an efficient way is to
> let the FileSystem implementation handle the seek behaviour underneath the
> API to be the most efficient as possible.
> A better approach to the seek problem is to provide a sequence of read
> locations as part of a single call, while letting the system schedule/plan
> the reads ahead of time.
> This is exceedingly useful for seek-heavy readers on HDFS, since this allows
> for potentially optimizing away the seek-gaps within the FSDataInputStream
> implementation.
> For seek+read systems with even more latency than locally-attached disks,
> something like a {{readFully(long[] offsets, ByteBuffer[] chunks)}} would
> take of the seeks internally while reading chunk.remaining() bytes into each
> chunk (which may be {{slice()}}ed off a bigger buffer).
> The base implementation can stub in this as a sequence of seeks + read() into
> ByteBuffers, without forcing each FS implementation to override this in any
> way.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
