[rpd] Re: Factors affecting in-region utilization - way forward?

[Mukom Akong T.]

On Mon, Jul 21, 2014 at 12:38 AM, Mark Tinka <mark.tinka at seacom.mu> wrote:

> On Sunday, July 20, 2014 10:19:04 PM Mukom Akong T. wrote:
>
> > This isn't a valid scenario. It's merely a factor that
> > feeds into the second scenario.
>
> Maintaining translation state on a device that cannot scale
> to handle that state will crash due to lack of memory.
>
> NAT'ing the software path will increase CPU utilization,
> which will slow down performance.
>
> These are real problems.
>


Correct. And these are scenario b related.  I read Seun's statement "...NAT
affecting performance on bandwidth increase" to mean that bandwidth
increase in itself will somehow worsen NAT performance. My response was to
the fact that it doesn't directly do so, instead it influences users
behavior so that they start doing more 'sophisticated' things what require
more ports and thus more entries in translation table which is the scenario
you quite correctly describe above.



>
> Mark.
>



-- 

Mukom Akong T.

http://about.me/perfexcellence |  twitter: @perfexcellent
------------------------------------------------------------------------------------------------------------------------------------------
“When you work, you are the FLUTE through whose lungs the whispering of the
hours turns to MUSIC" - Kahlil Gibran
-------------------------------------------------------------------------------------------------------------------------------------------
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.afrinic.net/pipermail/rpd/attachments/20140721/b34fa26a/attachment.html>