Jon,
I likely introduced some confusion into this by misnaming the IP option
I meant to refer to
when I said "IP explicit routing" - I meant "IP Source Routing" which provides
a list of IP routers to be
used in forwarding an IP packet.
In the past, this has been regarded by most as anathema because it
involves manipulating
IP header information in ways that few (if any) routers are able to handle in
the dataplane.
Not sure what you mean by "IP explicit (not static) routing methods."
Weighting of ECMP paths is a different thing entirely.
Sorry about the confusion...
--
Eric
-----Original Message-----
From: Jon Mitchell [mailto:[email protected]]
Sent: Thursday, June 06, 2013 8:49 PM
To: Eric Gray
Cc: Linda Dunbar; [email protected]; Lucy yong
Subject: Re: [nvo3] Is there any difference between "Network Virtualization
Overlay" and "Network Overlay"?
Importance: High
On 04/06/13 20:53 +0000, Eric Gray wrote:
> I don't believe we have yet eliminated the potential need for traffic
> engineered paths in virtual networks used for data-centers. And I
> doubt anyone is willing to consider using IP explicit routing to accomplish
> this, in the even that it is needed.
Generally agree with your overall sentiment but just a nit - some are willing
to consider using IP explicit (not static) routing methods (including just
weighting by multiple path introduction) to acheive non-MPLS based TE in DC's.
Jon
_______________________________________________
nvo3 mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/nvo3
