Robert,

Please see inline.

> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf Of Robert
> Raszuk
> Sent: Friday, December 21, 2012 2:18 PM
> To: Lucy yong
> Cc: Yakov Rekhter; Thomas Narten; Kireeti Kompella; Aldrin Isaac;
> NAPIERALA, MARIA H; [email protected]
> Subject: Re: [nvo3] Multi-subnet VNs [was Re: FW: New Version
> Notification for draft-yong-nvo3-frwk-dpreq-addition-00.txt]
> 
> Hi,
> 
> > [Lucy] In your DC, do VM or physical server support Pedro's solution?
> if
> > not, how do you run rfc4364 in your DC?
> 
> I am not sure what you are specifically referring to as "Pedro's
> solution"
[Lucy] something we have common reference. 
> 
> L3vpn for end systems is useful in decoupling control and data plane -
> so if you are asking for that - the answer is yes.
[Lucy] thanks.
> 
> However there are many ways to accomplish the same decoupling or even
> I am aware about work of some vendor to run BGP native to the hosts.
[Lucy] I can't reference this as rfc4364 that, imo, couple control and data 
plane.
> 
> I think to organize the discussion we should be very specific and talk
> SAFI here.
> 
> SAFI 128 is useful to interconnect IP subnets. SAFI 70 may be useful
> to interconnect L2 segments.
[Lucy] That is clear.
> 
> 
> > [Lucy] I do not say that E-IP-VPN for everything. we have EVPN and IP
> > VPN.
> 
> As the name suggestes "E-IP-VPN" does integrate both EVPN and IPVPN.
> What else it does not cover ? Token Ring or FDDI L2 ?
[Lucy] you have applications that fit better for EVPN or IPVPN. You also have 
the applications that 
require both. You can combine them as you want. However, some operator cost is 
expensive. For simple vanilla application, operation cost saving is beneficial. 
  
> 
> If you observe this thread carefully which apparently even Yakov have
> missed the claim Kireeti has made is very clear what purpose E-IP-VPN
> is to serve. Quote:
> 
> "b) In the quite common case where all traffic from a TS is IP, you
> don't have to maintain two tables and two forwarding paradigms at the
> NVE (one for IPs and one for MACs).  This is common enough to warrant
> optimization."
[Lucy] This is implementation specify. It is not necessary to maintain two 
tables.

Lucy 
> 
> > [Lucy] where was the original idea? you mean to let operator deal
> with
> > rfc4364 and evpn for their applications?
> 
> Why not ?
> 
> Cheers,
> R.
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3

Reply via email to