Re: [sop] [Sdnp] Fwd: SOP and SDN Question

One of the questions I had on an earlier thread was a comparison /
contrast between the Application <--> Controller <--> Infrastructure
model versus the Controller <--> {Application / Infrastructure} model.
As an example, what happens if the application that is supposed to
provision the network for its purposes fails (because of software or
underlying infrastructure failures)? Who bootstraps the infrastructure
and then the application?

The second issue I wonder is the nature of the control SDN (D being
whatever it is) wants to exercise over the network. As an example, one
blog recently talked about why OF not good for solving the distributed
forwarding state problem, but good solution for solving global
optimization issues.

http://networkheresy.wordpress.com/2011/11/17/is-openflowsdn-good-at-for
warding/

What stand does SDNP take regarding this?

Thanks, Ashish

-----Original Message-----
From: sop-bounces@ietf.org [mailto:sop-bounces@ietf.org] On Behalf Of
Sam Aldrin
Sent: Friday, February 24, 2012 9:39 PM
To: Thomas Nadeau
Cc: sdnp@lucidvision.com; sop@ietf.org; robert@raszuk.net
Subject: Re: [sop] [Sdnp] Fwd: SOP and SDN Question

I think SDN is being over loaded and confusion exists, not just between
Driven and Defined, but also what each of those does. The way i
understood and also being presented is, ONF SDN encompasses not just
openflow, but network applications with the help of northbound API
between controller and applications. Where SDNP plays pivotal role in
defining protocol elements  and api to interconnect network elements and
applications. It is high time to gets the terms identified along with
clear definition of what it does, and more importantly, start using them
in future conversations could alleviate the confusion.

Cheers
Sam

Sent from my iPad

On Feb 24, 2012, at 7:48 AM, Thomas Nadeau <tnadeau@lucidvision.com>
wrote:

> 
> On Feb 24, 2012, at 10:43 AM, Robert Raszuk wrote:
> 
>> Hi Tom,
>> 
>>>> You seem to suggest that Open Flow can be expanded to include
services
>>>> besides networking.
>>> 
>>> I think you misunderstood what I wrote. Software Defined Networks
(i.e.:
>>> OpenFlow) is targeted at working at the hardware abstraction level
only.
>> 
>> That's completely inaccurate.
> 
>    Not completely. *)
>    
>> SDN is all about network applications. You perhaps are confusing it
with ONF which indeed at this point is focused on defining OpenFlow
protocol.
> 
>    Software DRIVEN Networks is all about network applications *and*
their interaction with network services and elements, of which a
> subset is Software *defined* Networks (I.e.: OpenFlow).   We really
need to pick a different term for Software Driven Networks because 
> it is confusing everyone. People seem to equate "SDN" with OpenFlow
without being clear about the "D".
> 
>    --Tom
> 
> 
> 
>> But protocols are not defined just by bunch of hardware enthusiasts.
They are defined to address specific software defined applications.
>> 
>> Best,
>> R.
>> 
>> 
>> 
>> Software *Driven* Networks is
>>> targeted at services and network elements.
>>> There is a clear difference.
>>> 
>>> --Tom
>> 
>> 
> 
> _______________________________________________
> SDNP mailing list
> SDNP@lucidvision.com
> http://lucidvision.com/mailman/listinfo/sdnp
_______________________________________________
sop mailing list
sop@ietf.org
https://www.ietf.org/mailman/listinfo/sop