[
https://issues.apache.org/jira/browse/HDFS-941?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
bc Wong updated HDFS-941:
-------------------------
Attachment: HDFS-941-1.patch
Attaching a patch, which:
* In {{DFSInputStream}}, stores the last socket and the owner ({{BlockReader}})
of that socket.
* If the owner did read till EOS, then it'll consider reusing that socket.
* It's a cache of size 1, but easy to extend to _n_. However, size of 1 makes
sense for now because:
* The DN ({{DataXceiver}}) no longer closes the client connection after every
op. It optimistically expects the next op, with a very low timeout value. It
closes the connection only on EOS or timeout.
Note that the HDFS-941-1.patch depends on the HDFS-1001 patch.
> Datanode xceiver protocol should allow reuse of a connection
> ------------------------------------------------------------
>
> Key: HDFS-941
> URL: https://issues.apache.org/jira/browse/HDFS-941
> Project: Hadoop HDFS
> Issue Type: Improvement
> Components: data-node, hdfs client
> Affects Versions: 0.22.0
> Reporter: Todd Lipcon
> Attachments: HDFS-941-1.patch
>
>
> Right now each connection into the datanode xceiver only processes one
> operation.
> In the case that an operation leaves the stream in a well-defined state (eg a
> client reads to the end of a block successfully) the same connection could be
> reused for a second operation. This should improve random read performance
> significantly.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
