RE: Advancing the Protocol and Morin Drafts

[Renwei Li <[e-mail]>]

Hi Dino,

Your point is crystal-clear!

To see whether or not a protocol or a technology is successful, we should
see whether or not it has solved a real problem and meet the demand from
customers. 

More than 10 years ago, for example, there was a demand for wireless
communications. Two technologies were competing: satellite vs. base station.
Satellite people said that a service provider would need to install
thousands, and even millions of, base stations for wireless communications,
while a provider would just need to launch a few tens, and at most a few
hundreds, of satellites. Millions of base stations mean high OPEX: they are
hard to configure, hard to manage, hard to upgrade... But look at now,
satellites are gone. Instead, millions of base stations are installed. Yes,
no doubt the total cost is high, but there has been an increasing number of
wireless subscribers, and thus the total cost is paid off.

Now return to the issue of BGP/MPLS VPN. Is there a decreasing number of
L3VPN customers? From what I know, there is still an increasing number of
L3VPN sites and VPN routes. This just means that such a technology is not
being replaced by others. This doesn't mean BGP has no problems. I admit
that BGP is not scaling well for L3VPN, but what else can we do? For L3VPN,
unless you want to write a new protocol, you have to tweak an existing one. 
Suppose that you decide to tweak a protocol, now what's the choice? Most of
the routing/multicasting/mpls protocols are basically hop by hop except
bgp/ldp/msdp, from which BGP looks a clear choice. 

Having said that about l3vpn, three types of VPNs co-exist now:
host-provided, CE-provided, and PE-provided. It doesn't seem that these
three can be replaced by each other, as they solve different real problems
in different contexts. 

Just add more spice to cook your topic -:)

Cheers,

Renwei



> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] On Behalf Of
> Dino Farinacci
> Sent: Saturday, October 11, 2008 6:12 PM
> To: Drake, John E
> Cc: [e-mail]; [e-mail]; [e-mail]
> Subject: Re: Advancing the Protocol and Morin Drafts
> 
> It should have been clear. And if it is was not, then you can't even
> come close to claiming success.
> 
> Dino
> 
> On Oct 11, 2008, at 12:55 PM, Drake, John E wrote:
> 
> > And your point is?
> >
> > John Drake
> > Boeing Satellite Systems
> > 2260 East Imperial Highway MC W-S05-P208
> > El Segundo, CA 90245
> > [e-mail]
> > [phone]
> >
> > ----- Original Message -----
> > From: Dino Farinacci <[e-mail]>
> > To: Yakov Rekhter <[e-mail]>
> > Cc: Ross Callon <[e-mail]>; [e-mail] <[e-mail]>
> > Sent: Fri Oct 10 16:35:11 2008
> > Subject: Re: Advancing the Protocol and Morin Drafts
> >
> > Define success Yakov?
> >
> > It is not at all clear, and has never been proved to me this service
> > is *profitable*. It may be generating revenue but the cost of running
> > the service must be considered as well.
> >
> > Dino
> >
> > On Oct 10, 2008, at 9:08 AM, Yakov Rekhter wrote:
> >
> >> Prasanna,
> >>
> >>> I vote NO.
> >>>
> >>> Being in AT&T Advanced Tier support group and having supported the
> >>> MVPN
> >>> core for the last 2 years, I can clearly say that we need a solution
> >>> such as PIM-BiDir that can reduce the number of multicast states in
> >>> the
> >>> PEs to be able to scale the Groups X Sources x OILs explosion for
> >>> our
> >>> very large Enterprise customers.   We definitely do not want to
> >>> tweak a
> >>> crucial protocol like BGP of which our scale is 2 Million VPNV4
> >>> routes
> >>> in the US alone.
> >>
> >> On the subject of "we definitely do not want to tweak", I'd like
> >> to remind you that during the early days of 2547 VPNs some of its
> >> opponents were saying that they are against 2547 VPNs because they
> >> do not want to "tweak" such a crucial protocol as BGP to carry VPNv4
> >> routes.
> >>
> >> Today AT&T has "2 million VPNv4 routes in the US alone" all carried
> >> in BGP, and a successful 2547 VPN service. None of this would be
> >> possible if we would not tweak BGP.
> >>
> >> Yakov.
> >