[
https://issues.apache.org/jira/browse/HDFS-4661?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13623334#comment-13623334
]
Tsz Wo (Nicholas), SZE commented on HDFS-4661:
----------------------------------------------
- We should not add requestShortCircuitFds to DataTransferProtocol since
DataTransferProtocol is for transferring blocks or large amount of data. We
should add it to ClientDatanodeProtocol.
- It is not a good idea to pass caches to BlockReader.close(..). BlockReader
is an interface. The implementations should take care the cache implicitly.
- There are quite a few configuration properties added to the branch but there
is only a very simple example configuration in the documentation. Please
document all the short circuit related configuration properties.
> code style fixes suggested by Nicholas
> --------------------------------------
>
> Key: HDFS-4661
> URL: https://issues.apache.org/jira/browse/HDFS-4661
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: datanode, hdfs-client, performance
> Reporter: Colin Patrick McCabe
> Priority: Minor
> Attachments: HDFS-4661.001.patch, HDFS-4661.002.patch
>
>
> * The log statement in DataXceiver
> "BlockSender.ClientTraceLog.info("REQUEST_SHORT_CIRCUIT_FDS"...)" could be
> cleaned up somewhat.
> * use {{FsDatasetSpi#getBlockInputStream}} and
> {{FsDatasetSpi#getMetaDataInputStream}} rather than adding
> {{FsDatasetSpi#getShortCircuitFdsForRead}}
> * {{FileInputStreamCache.Key.equals}}: use short-circuit boolean AND
> * In FileInputStreamCache.CacheCleaner, the code "iter =
> map.entries().iterator()" can be removed with the same result since the
> (previous) first element must be removed.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
