[
https://issues.apache.org/jira/browse/HDFS-8319?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14559029#comment-14559029
]
Walter Su commented on HDFS-8319:
---------------------------------
9. Pread create block reader at {{actualGetFromOneDataNode}}. It's a problem of
PositionStripeReader version's {{readChunk()}}. It creates too much readers.
10.
For trunk branch's DFSInputStream:
stateful read may seek to next block. It uses BlockSeekTo() to handle error.
pread handle error at {{actualGetFromOneDataNode}}, it uses getBlockRange() to
locate block.
The difference is that stateful read seek to next block while reading.
For EC branch's StripedDFSInputStream:
We have no such difference at {{readChunk()}}. You have two version of
{{readChunk()}}. We only have to keep one of them.
> Erasure Coding: support decoding for stateful read
> --------------------------------------------------
>
> Key: HDFS-8319
> URL: https://issues.apache.org/jira/browse/HDFS-8319
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Reporter: Jing Zhao
> Assignee: Jing Zhao
> Attachments: HDFS-8319.001.patch
>
>
> HDFS-7678 adds the decoding functionality for pread. This jira plans to add
> decoding to stateful read.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
