You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hc.apache.org by "ivan lappo (Created) (JIRA)" <ji...@apache.org> on 2011/12/26 17:44:32 UTC

[jira] [Created] (HTTPCORE-287) HTTP.NIO Performance problems

HTTP.NIO Performance problems
-----------------------------

                 Key: HTTPCORE-287
                 URL: https://issues.apache.org/jira/browse/HTTPCORE-287
             Project: HttpComponents HttpCore
          Issue Type: Task
          Components: HttpCore NIO
    Affects Versions: 4.1.1, 4.1.4
            Reporter: ivan lappo


I use DefaultConnectiongIOReactor to request connection. When i receive connection i submit http request for it. After receiving responce, i release it for new requets. I constantly increase number of requests executed per minute
on the chart there are few params shown with discretisation 15 seconds - green - number of requests executed, red - average responce time (between connection.requestOutput() method and received responce, sky-blue - number of requests that failed to execute because of connection being closed). Horisontal axis represents timeline, Left vertical axis - values of these charts. I monitored state of my machine using mvstat and it shows for first 10 minutes 4-10 % CPU, 35-40 after. This was tested on m1.small amazon instance. With c1.medium (which has 4 times more computation CPU power) it behaves the same, but with upper bound of 16000 requests per 15 seconds (against 5000 on weeker machine) - therefore it is not networking problem. 
What can cause such behaviour, and what can i do to improve performance? 
(http://www.freeimagehosting.net/eba15) - link to image

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org

[jira] [Resolved] (HTTPCORE-287) HTTP.NIO Performance problems

Posted by "Oleg Kalnichevski (Resolved) (JIRA)" <ji...@apache.org>.
     [ https://issues.apache.org/jira/browse/HTTPCORE-287?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Oleg Kalnichevski resolved HTTPCORE-287.
----------------------------------------

    Resolution: Invalid

Ivan,

Jira is an issue tracker and not a support forum. Please post your questions to the HC dev or user list

http://hc.apache.org/mail-lists.html 

PS: honestly, I am not sure I understand the problem you are having. Besides, if you want an HTTP client based on HttpCore, you might be better off using HttpAsyncClient instead of developing one of your own.

http://hc.apache.org/httpcomponents-asyncclient-dev/index.html

Oleg
                
> HTTP.NIO Performance problems
> -----------------------------
>
>                 Key: HTTPCORE-287
>                 URL: https://issues.apache.org/jira/browse/HTTPCORE-287
>             Project: HttpComponents HttpCore
>          Issue Type: Task
>          Components: HttpCore NIO
>    Affects Versions: 4.1.1, 4.1.4
>            Reporter: ivan lappo
>
> I use DefaultConnectiongIOReactor to request connection. When i receive connection i submit http request for it. After receiving responce, i release it for new requets. I constantly increase number of requests executed per minute
> on the chart there are few params shown with discretisation 15 seconds - green - number of requests executed, red - average responce time (between connection.requestOutput() method and received responce, sky-blue - number of requests that failed to execute because of connection being closed). Horisontal axis represents timeline, Left vertical axis - values of these charts. I monitored state of my machine using mvstat and it shows for first 10 minutes 4-10 % CPU, 35-40 after. This was tested on m1.small amazon instance. With c1.medium (which has 4 times more computation CPU power) it behaves the same, but with upper bound of 16000 requests per 15 seconds (against 5000 on weeker machine) - therefore it is not networking problem. 
> What can cause such behaviour, and what can i do to improve performance? 
> (http://www.freeimagehosting.net/eba15) - link to image

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@hc.apache.org
For additional commands, e-mail: dev-help@hc.apache.org