[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [RRG] Re: [RAM] Tunnelling & GigE jumbo frame size



On Wed, Sep 12, 2007 at 06:53:26PM -0700, Dino Farinacci wrote:
> >Where is this assumption coming from?  As long as there are FastE  
> >links,
> >this is not going to happen.
> 
> Listen folks, one solution isn't going to solve all of the world's  
> problems. Want to throw away this effort because of this one issue?  
> You think another solution won't have issues as well?

	I have to agree with Dino on this one. What we are
	talking about here is engineering. And more than that,
	its engineering with a large constraint set (i.e., the
	Internet). But engineering is about tradeoffs (well, its
	part art, part economics, and part luck too). So as Dino
	says, other solutions are going to have different
	tradeoffs. Some worse (for sure), and possibly some
	better. But it is clear that all of the solutions on the
	table (and likely any other solution) have complex issues
	sets, so we need to look at the bigger picture (rather
	than focus on any one issue in isolation).

> We design for the optimal case and hopefully with experience and data  
> from operators more networks move to the optimal case. 

	Right, and experience and data are the fundemental
	compliments to both research and engineering.  

> And later if  we have to still solve the suboptimal case, we do
> fragmentation. 
>
> Can we agree with this?

	Really.

	Dave

Attachment: signature.asc
Description: Digital signature