Last Mile Design

Brandon Martin lists.nanog at
Fri Feb 8 17:44:18 UTC 2019

On 2/8/19 2:07 AM, Mark Tinka wrote:
>> I do like to separate SMB and Resi traffic, but it's mostly for
>> customer service reasons rather than technical reasons.  That
>> separation rarely entails separate equipment but rather just VLANs and
>> PCPs, IP subnets, etc.
> Many years ago, I did consider running both Consumer and Enterprise
> traffic on one router - and for purposes of pride, I'm sure the major
> vendors would like to boast that they could allow you to do this. But in
> practice, it's probably a bad idea... BNG's have too many moving parts,
> and for some platforms, there is actually special code optimized for BNG
> deployments that may have an impact on traditional Enterprise or Service
> Provider customers.
> So I would separate BNG's from regular edge routers.

Enterprise DIA is a whole different beast.  For sure, that stays 
separate at least for now.  Some of the forthcoming PON technologies 
have so much bandwidth that it may become attractive to start merging 
them at the access layer for smaller customers, and then I guess we'll 
have to see what the best way to handle L3 termination on that is.

If anything, just ensuring that the (often) separate tech teams have the 
proper access to it and knowledge of what the others are doing might be 
a bit of an issue.

>> If you're in a position where you want to or have to offer competitors
>> access to your network to sell service directly to customers, that's
>> also going to potentially really change the situation.
> Why? Chances are they will require Ethernet access between their
> customer and their head-end, which is a typical scenario.

I'm thinking that, if you push L3 termination all the way out to the 
last access node (FTTN DSLAM being the obvious one here), you may then 
lack a decent way to haul pure Ethernet back to their head-end.  If your 
L3 termination also supports MPLS, or Q-in-Q, you're probably fine.  The 
latter might negate the potential advantages of distributed L3 from a 
routing POV by forcing you to again run STP or similar.

If you're doing L3 termination a bit more centralized, even if not with 
big behemoths on a "one per super-metro" basis, this may not be a 
problem at all.  HFC and FTTx PONs might end up being like that 
inherently just because of the nature of the plant and tech that runs on it.
Brandon Martin

More information about the NANOG mailing list