[
https://issues.apache.org/jira/browse/HADOOP-13144?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16391669#comment-16391669
]
Íñigo Goiri commented on HADOOP-13144:
--------------------------------------
Thanks [~ywskycn] for trying this out; I posted [^HADOOP-13144.001.patch] with
the fixes for compilation.
I submitted the patch so Yetus should cover the next ones.
In general, I think this is touching a pretty sensitive part of the Hadoop code
but I think the modifications are pretty minimal.
At the same time, as [~ywskycn] pointed out, it helps dramatically with the
performance of the Routers for HDFS.
We would open a separate JIRA for the Router connection creation if this goes
in.
Anybody available for a review?
> Enhancing IPC client throughput via multiple connections per user
> -----------------------------------------------------------------
>
> Key: HADOOP-13144
> URL: https://issues.apache.org/jira/browse/HADOOP-13144
> Project: Hadoop Common
> Issue Type: Improvement
> Components: ipc
> Reporter: Jason Kace
> Assignee: Íñigo Goiri
> Priority: Minor
> Attachments: HADOOP-13144.000.patch, HADOOP-13144.001.patch
>
>
> The generic IPC client ({{org.apache.hadoop.ipc.Client}}) utilizes a single
> connection thread for each {{ConnectionId}}. The {{ConnectionId}} is unique
> to the connection's remote address, ticket and protocol. Each ConnectionId
> is 1:1 mapped to a connection thread by the client via a map cache.
> The result is to serialize all IPC read/write activity through a single
> thread for a each user/ticket + address. If a single user makes repeated
> calls (1k-100k/sec) to the same destination, the IPC client becomes a
> bottleneck.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
