[nvrg-bof] [Fwd: Call for agenda items for the session at the 73rd IETF meeting]

[Martin Stiemerling <martin.stiemerling at cs.uni-goettingen.de>]

Hi,

This is a reminder to send me agenda items.

right now there is one presentation planned out of the EU IST FP7 4WARD 
project.

Let me know if you interested in giving a talk.

Martin   

-------- Original Message --------
Subject: 	[nvrg-bof] Call for agenda items for the session at the 73rd 
IETF meeting
Date: 	Fri, 7 Nov 2008 15:29:00 +0100
From: 	Martin Stiemerling <Stiemerling at nw.neclab.eu>
To: 	<nvrg at listserv.gwdg.de>



Dear all,

We have the opportunity to meet during the IETF#73 meeting in Minneapolis:

Thursday, 1740-1930 Afternoon Session III
(see under reservations for IESG breakout room here:
http://trac.tools.ietf.org/group/iesg/trac/wiki/UpcomingIetf)


The purpose of the first meeting in March 2008 was to get in contact and
getting an overview of existing works and projects. However, now it is
time to consider the research challenges to be solved by such a possible
Network Virtualization Research Group. 

Network virtualization seems to be a broad term (ranging from optical up
to the application layer) and getting started on discussing research
challenges requires an outline on what people might find challenging. 
Some technology is already available that could be used to virtualized
networks, such as, but not limited to, multiple lambdas on a fibre, the
Virtual Internet Architecture (VIA), and the current trend in virtual
routers (e.g., reusing virtual machine technology). 

There are already existing technologies for network virtualization and
in various cases they are used. For instance, in EMULAB or VINI
virtualization is used for testbed operations. Management of these
virtual networks might be relatively easy (they are not - but a bit
easier than what comes now) compared to large-scale virtual network
that spans the whole Internet and that is not under a single
administrative control. Management in network virtualization will be
needed in various parts, such as, management of the substrate,
management of each slice in the substrate, and probably management of
the combination of substrate with an arbitrary number of slices.

The current trend of virtualization on devices is much focussed on
router virtualization, but what about the virtualization on the end
hosts? Yes, there is the virtual machine concept, but is this the right
approach or is it better to go for stack virtualization? 
(an example for stack virtualization is here:
http://imunes.tel.fer.hr/virtnet/)


If you have a research challenge and want to discuss it (or them), please
drop me a mail until Wednesday, Nov 12th. A presentation is appreciated. 

The very rough agenda:
- introduction
- (technical) presentations
- discussion of challenges
- charter discussion


At the end of the session, the attendees can discuss the charter of a
possible NVRG. A RG does need a charter and Aaron (as IRTF chair) urged
me to get this progressed in terms of stating what the scope of this RG
is and issues are to be solved.

Let me and the others know your thoughts.

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