[
https://issues.apache.org/jira/browse/HADOOP-6433?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12790043#action_12790043
]
Zheng Shao commented on HADOOP-6433:
------------------------------------
AsyncDiskService is designed to work with multiple disk volumes, to make sure
we have a maximum number of threads per volume.
Both the use cases in HDFS and Mapreduce are like that.
I don't think it will be used outside this area in the near future - for
non-multi-volume disk operations, people will probably use a single thread pool
instead of AsyncDiskService (which contains multiple thread pools)
So I feel it's good to name it AsyncDiskService instead of AsyncService.
> Add AsyncDiskService that is used in both hdfs and mapreduce
> ------------------------------------------------------------
>
> Key: HADOOP-6433
> URL: https://issues.apache.org/jira/browse/HADOOP-6433
> Project: Hadoop Common
> Issue Type: New Feature
> Reporter: Zheng Shao
> Assignee: Zheng Shao
> Attachments: HADOOP-6433.1.patch, HADOOP-6433.2.patch,
> HADOOP-6433.3.patch
>
>
> Both MAPREDUCE-1213 and HDFS-611 are using a class called AsyncDiskService.
> The idea is to create a thread pool per disk volume, and use that for
> scheduling async disk operations.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
