[
https://issues.apache.org/jira/browse/HDFS-2834?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13233095#comment-13233095
]
[email protected] commented on HDFS-2834:
-----------------------------------------------------
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/4212/#review6103
-----------------------------------------------------------
Real close now!
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/BlockReaderLocal.java
<https://reviews.apache.org/r/4212/#comment13128>
this comment seems like it's in the wrong spot, since the code that comes
after it doesn't reference offsetFromChunkBoundary.
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestBlockReaderLocal.java
<https://reviews.apache.org/r/4212/#comment13130>
shouldn't this be true?
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestBlockReaderLocal.java
<https://reviews.apache.org/r/4212/#comment13132>
no reason to use DFSClient here. Instead you can just use the filesystem,
right? Then downcast the stream you get back?
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestBlockReaderLocal.java
<https://reviews.apache.org/r/4212/#comment13131>
don't you want an assert on sawException here? You can also use
GenericTestUtils.assertExceptionContains() if you want to check the text of it
- Todd
On 2012-03-09 00:47:24, Henry Robinson wrote:
bq.
bq. -----------------------------------------------------------
bq. This is an automatically generated e-mail. To reply, visit:
bq. https://reviews.apache.org/r/4212/
bq. -----------------------------------------------------------
bq.
bq. (Updated 2012-03-09 00:47:24)
bq.
bq.
bq. Review request for hadoop-hdfs and Todd Lipcon.
bq.
bq.
bq. Summary
bq. -------
bq.
bq. New patch for HDFS-2834 (I can't update the old review request).
bq.
bq.
bq. This addresses bug HDFS-2834.
bq. http://issues.apache.org/jira/browse/HDFS-2834
bq.
bq.
bq. Diffs
bq. -----
bq.
bq.
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/BlockReader.java
dfab730
bq.
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/BlockReaderLocal.java
cc61697
bq.
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSConfigKeys.java
4187f1c
bq.
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/DFSInputStream.java
2b817ff
bq.
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/RemoteBlockReader.java
b7da8d4
bq.
hadoop-hdfs-project/hadoop-hdfs/src/main/java/org/apache/hadoop/hdfs/RemoteBlockReader2.java
ea24777
bq.
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/BlockReaderTestUtil.java
9d4f4a2
bq.
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestBlockReaderLocal.java
PRE-CREATION
bq.
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestParallelRead.java
bbd0012
bq.
hadoop-hdfs-project/hadoop-hdfs/src/test/java/org/apache/hadoop/hdfs/TestShortCircuitLocalRead.java
eb2a1d8
bq.
bq. Diff: https://reviews.apache.org/r/4212/diff
bq.
bq.
bq. Testing
bq. -------
bq.
bq.
bq. Thanks,
bq.
bq. Henry
bq.
bq.
> ByteBuffer-based read API for DFSInputStream
> --------------------------------------------
>
> Key: HDFS-2834
> URL: https://issues.apache.org/jira/browse/HDFS-2834
> Project: Hadoop HDFS
> Issue Type: Improvement
> Reporter: Henry Robinson
> Assignee: Henry Robinson
> Attachments: HDFS-2834-no-common.patch, HDFS-2834.3.patch,
> HDFS-2834.4.patch, HDFS-2834.5.patch, HDFS-2834.6.patch, HDFS-2834.7.patch,
> HDFS-2834.8.patch, HDFS-2834.9.patch, HDFS-2834.patch, HDFS-2834.patch,
> hdfs-2834-libhdfs-benchmark.png
>
>
> The {{DFSInputStream}} read-path always copies bytes into a JVM-allocated
> {{byte[]}}. Although for many clients this is desired behaviour, in certain
> situations, such as native-reads through libhdfs, this imposes an extra copy
> penalty since the {{byte[]}} needs to be copied out again into a natively
> readable memory area.
> For these cases, it would be preferable to allow the client to supply its own
> buffer, wrapped in a {{ByteBuffer}}, to avoid that final copy overhead.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
