[
https://issues.apache.org/jira/browse/HDFS-2129?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Todd Lipcon updated HDFS-2129:
------------------------------
Attachment: hdfs-2129-benchmark.png
hdfs-2129.txt
Attaching updated patch and benchmark.
The benchmark is from running TestParallelRead ten times each with the patch
and without the patch, and throwing away the first two runs to ignore JVM
warmup. The patch provides a modest improvement for 4 and 8 concurrent reader
threads, and no significant difference for 16. This makes sense since the test
machine has only 4 physical cores plus hyperthreading.
> Simplify BlockReader to not inherit from FSInputChecker
> -------------------------------------------------------
>
> Key: HDFS-2129
> URL: https://issues.apache.org/jira/browse/HDFS-2129
> Project: Hadoop HDFS
> Issue Type: Sub-task
> Components: hdfs client
> Reporter: Todd Lipcon
> Assignee: Todd Lipcon
> Fix For: 0.23.0
>
> Attachments: hdfs-2129-benchmark.png, hdfs-2129.txt, hdfs-2129.txt
>
>
> BlockReader is currently quite complicated since it has to conform to the
> FSInputChecker inheritance structure. It would be much simpler to implement
> it standalone. Benchmarking indicates it's slightly faster, as well.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
