Re: [nvrg-bof] What is Network Virtualization?

["Martin Stiemerling" <Stiemerling at nw.neclab.eu>]

Joe, all,

> -----Original Message-----
> From: nvrg-bounces at listserv.gwdg.de 
> [mailto:nvrg-bounces at listserv.gwdg.de] On Behalf Of Joe Touch
> Sent: Tuesday, June 17, 2008 7:11 PM
> To: Martin Stiemerling
> Cc: nvrg at listserv.gwdg.de
> Subject: Re: [nvrg-bof] What is Network Virtualization?
> 
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> Martin (et al.),
> 
> Martin Stiemerling wrote:
> | Hi all,
> |
> | It has been quite silent on the NVRG list for a while. 
> Sorry for that.
> |
> | I have been stepping back a bit and considered what network
> virtualization actually might be.
> |
> | So is there the first attempt to get some discussions started.
> |
> | Apparently there are different camps in the area of network
> virtualization (at least from my recent experience):
> | a) network virtualization in test beds (GENI, parts of 
> 4WARD, Akari, 
> | etc)
> 	while interesting, this seems an application of (b) below,
> 	and decidedly uninteresting architecturally; testbeds have no
> 	unique requirements in this area, IMO

I also see that a) is an application of b) but time wise (at least right now) a) is hot. Even though I personally tend also to b) for the architectural aspects. 

> 	I'd include rbridges/trill and L2VPNs here.

Ok.

> 
> | b) network virtualization for the Internet (parts of 4WARD, my own
> idea, etc)
> 	including X-Bone, FWIW ;-)

Yep, forgot that when writing. Thanks for reminding! :)

> 
> | c) existing parts, for instance, l*vpn in the IETF and 
> virtual router
> support in existing devices.
> 	this includes LISP and NERD, but I would not include L2VPNs
> 	or L3VPNs here.
> 
> (a) arguably includes virtualization at any layer
> (c) arguably are components of (b)

Agreed. 

> 
> IMO, the IRTF/IETF ought to focus on (b) and the components 
> of (c) that support (b).

I do not want to rule out a) if people on the list feel that this is the right place to discuss issues related to this. However, there can be other venues better suitable for discussing this, e.g., GENI. 

Personally, I also see the focus on b) and components of c) that support b). But as you've said a) might be special case of b).

Thanks,

  Martin 


stiemerling at nw.neclab.eu

NEC Laboratories Europe - Network Research Division
NEC Europe Limited | Registered Office: NEC House, 1 Victoria Road, London W3 6BL | Registered in England 2832014