hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Chris Nauroth (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HADOOP-9691) RPC clients can generate call ID using AtomicInteger instead of synchronizing on the Client instance.
Date Wed, 03 Jul 2013 20:29:20 GMT

     [ https://issues.apache.org/jira/browse/HADOOP-9691?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Chris Nauroth updated HADOOP-9691:
----------------------------------

    Attachment: HADOOP-9691.1.patch

Attaching a patch to switch to using {{AtomicInteger}}.  I've also added a test that makes
RPC calls from multiple threads sharing the same client instance and then checks that every
call had a unique call ID in the expected sequential range.  This test shows the same results
before and after the switch to {{AtomicInteger}}.

To support the test, I needed to add a {{getCallId}} method to the server to expose call ID
from the current call in its thread-local storage.  I assume we'll eventually need this method
anyway for HDFS-4942 when we start attaching client ID + call ID to edit log ops.
                
> RPC clients can generate call ID using AtomicInteger instead of synchronizing on the
Client instance.
> -----------------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-9691
>                 URL: https://issues.apache.org/jira/browse/HADOOP-9691
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: ipc
>    Affects Versions: 3.0.0, 2.1.0-beta
>            Reporter: Chris Nauroth
>            Assignee: Chris Nauroth
>            Priority: Minor
>         Attachments: HADOOP-9691.1.patch
>
>
> As noted in discussion on HADOOP-9688, we can optimize generation of call ID in the RPC
client code.  Currently, it synchronizes on the {{Client}} instance to coordinate access to
a shared {{int}}.  We can switch this to {{AtomicInteger}} to avoid lock contention.

--
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

Mime
View raw message