[ofa-general] IPoIB-CM/RC - NAPI patch.

[Erez Strauss]

Hi Bernie,

 

Thank you for your reply.

 

In this case I'm using an application which is using very small messages
(300Bytes) and I can not change it.

The focus of the testing is to reduce the latency to minimum during high
stress, while keeping the application intact.

 

Similar synthetic test would be multiple 'iperf -l 300 ....' and sum of
the total bandwidth (or packets per seconds).

 

Any suggestions for IPoIB-CM/RC are welcome.

Do you expect the SDP to perform better under the above constrains?

 

Thanks,

Erez

Voltaire Inc.

 

________________________________

From: general-bounces at lists.openfabrics.org
[mailto:general-bounces at lists.openfabrics.org] On Behalf Of Bernard
King-Smith
Sent: Monday, March 05, 2007 6:27 PM
To: general at lists.openfabrics.org
Subject: Re: [ofa-general] IPoIB-CM/RC - NAPI patch.

 



> ----- Message from "Erez Strauss" <erezs at voltaire.com> on Tue, 6 Mar
> 2007 00:07:41 +0200 ----- 
> 
> To: 
> 
> <openib-general at lists.openfabrics.org> 
> 
> Subject: 
> 
> [ofa-general] IPoIB-CM/RC - NAPI patch. 
> 
> Hello Roland, 
>   
> I understand that you have a patch to enable NAPI for IPoIB-CM/RC. 
>   
> I'm interested in the patch to reduce the number of interrupts I get
> on a system (40k /s). 

Are you still getting 40k interrupts per second with IPoIB-CM? You
shouldn't because if you are using a 32768 MTU size you should cut the
interrupts down by a factor of 16. With IPoIB-CM, each interrupt should
be a 32K message instead of a 2K message. 

>   
> Does it apply to the nightly build? 
>   
> Are there tunable variables to control the increase of latency due 
> to reduction of interrupts? 
>   
> Would you please send me a copy? 
>   
> Thanks, 
> Erez 
> Voltaire Inc.



Bernie King-Smith  
IBM Corporation
Server Group
Cluster System Performance  
wombat2 at us.ibm.com    (845)433-8483
Tie. 293-8483 or wombat2 on NOTES 

"We are not responsible for the world we are born into, only for the
world we leave when we die.
So we have to accept what has gone before us and work to change the only
thing we can,
-- The Future." William Shatner

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/general/attachments/20070306/2c98b81e/attachment.html>