[
https://issues.apache.org/jira/browse/DIRMINA-461?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
falconhand updated DIRMINA-461:
-------------------------------
Description:
Hello,
There is a Mr. Michael T. Nygard explaining a TCP stall problem he met when
dealing with java NIO:
http://www.highscalability.com/comment/reply/168/360
The main idea is to read as much data as possible when a channel is selected as
readable to minimize the data cumulation in this socket's TCP buffer during
thread scheduling.
I haven't reproduced or verified this issue yet, however, I wonder is it a
potential threat to MINA as well?
Thanks.
was:
Hello,
There is a Mr. Michael T. Nygard explaining a TCP stall problem he met when
dealing with java NIO:
http://www.highscalability.com/comment/reply/168/360
The main idea is to read as much data as possible when a channel is selected as
readable to minimize the data cumulation during thread scheduling.
I haven't reproduced or verified this issue yet, however, I wonder is it a
potential threat to MINA as well?
Thanks.
> Possible TCP stall consideration
> --------------------------------
>
> Key: DIRMINA-461
> URL: https://issues.apache.org/jira/browse/DIRMINA-461
> Project: MINA
> Issue Type: Improvement
> Components: Core
> Reporter: falconhand
>
> Hello,
> There is a Mr. Michael T. Nygard explaining a TCP stall problem he met when
> dealing with java NIO:
> http://www.highscalability.com/comment/reply/168/360
> The main idea is to read as much data as possible when a channel is selected
> as readable to minimize the data cumulation in this socket's TCP buffer
> during thread scheduling.
> I haven't reproduced or verified this issue yet, however, I wonder is it a
> potential threat to MINA as well?
> Thanks.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
