Call for participation for possible IRTF Network Virtualization Research Group
UPDATE: Meeting room and agenda have been updated on 2008-03-11.
If you are interested you can
- visit this web site http://www.net.informatik.uni-goettingen.de/nvrg/ and
- subscribe to the discussion mailing list at https://listserv.gwdg.de/mailman/listinfo/nvrg.
The mailing list's address is nvrg@listserv.gwdg.de
The meeting at the IETF (BAR BOF style in IETF speak) is on
WEDNESDAY, March 12, 2008
1510-1700 Afternoon Session II and beyond
Meeting room: 415 (IAB breakout room)
How do I find this room? Go to the IETF terminal room, pass it and go left. There you find the 415 meeting room.
Agenda
* NOTE WELL (1 minute)
* Agenda Bashing (2 minutes)
* Intention of this meeting - Martin Stiemerling (5 minutes)
* Network Virtualization - Joe Touch (15 minutes + 5 minutes questions)
* Network Virtualization in the European 4ward research project - Roland Bless
(proxied by Martin Stiemerling, 15 minutes + 5 minutes questions)
* Network Virtualization Project at NICT - Shu Yamamoto (15 minutes + 5 minutes questions)
* Open Discussions (remaining time)
* Conclusion (10 minutes)
Several research initiatives in the area of “Future
Internet” have recently started in the US, Europe, and Asia.
Amongst them are, for instance, GENI/FIND [1], 4WARD [2], Trilogy [3],
and NICT's New Generation Network projects [4]. All these efforts
possible lead to an internetworking architecture with the support of
network virtualization that significantly provides more flexible
features than the current Internet.
Network virtualization follows the same principle as any known computer
virtualization technology, such as virtual memory, virtual hard disk,
virtual screens, etc. It is always an abstraction from a real resource,
hiding the underlying complexity (cf. [5]). But even if virtualization
is an integral part of today’s computing technology (which is
close to communication technology), it is still lacking a counterpart
in the Internet architecture. Current proposals discuss possible
virtualization techniques (e.g. [7]) on top the today's Internet (also
past techniques such as MBONE and related), or more general on a
substrate [6]. The clean slate approach [9] also considers network
virtualization as an integral part of the future Internet.
However, all of these activities are bound to each local community, but
there is no larger, international discussion (apart from some smaller
meetings, e.g., between Ambient Network and GENI/FIND [10].
All of currently discussed ideas are somewhat heading towards different
directions (especially different between academia and industry
research). Thus it would be beneficial to have a venue for discussing
ideas w.r.t. network virtualization as a new means for the Internet
today, as transition path to new technologies, or as integral part of
the future Internet would be beneficial. Researchers from US, Europe,
and Asia could discuss what exactly network virtualization could bring
as benefit, how it is impacting the Internet, and what might be the
impact to the Internet architecture.
Therefore, I have in mind a venue where researchers can discuss their
ideas (w/o considering standardization/interfaces definitions yet) and
try to shape a way forward for future Internet activities in this
field. The Internet Research Task Force (IRTF) can be good place to
give this a home.
The intention of this email is to first gather community feedback about
whether researchers do find this idea of creating an IRTF research
group (RG) appealing enough and have energy enough to actively
participate in such a group.
The envisioned main goals of such a RG in the area of network virtualization can be:
- a discussion venue without standardization or any other limiting boundaries
- bringing together the right people from US, Asia and Europe
Open issues to be addressed by this RG are, but not limited to:
- impact of network virtualization to the Internet architecture as
such, Internet devices (routers and hosts), and applications running in
the Internet;
- can network virtualization solve the ossification issue (cf. [6] of
the Internet or is it just shifting problems from one place to another;
- are overlays one possible answer to network virtualization;
- are the current testbed initiatives (e.g., PlanetLab, OneLab, etc) a first answer to network virtualization?
Interested people are requested to reply to this email and if possible
to meet at the upcoming IETF meeting (IETF#71) in Philadelphia to
discuss further steps. A possible meeting slot during the IETF meeting
would be Wednesday after the IETF Operations and Administration
Plenary. However, I'm open for suggestions.
References:
[1] GENI project web page, http://www.geni.net
[2] 4ward project page, http://ww.4ward-project.eu
[3] EU FP7 IST Trilogy project, http://www.trilogy-project.org
[4] NICT's NWGN projects, http://akari-project.nict.go.jp/document/INFOCOM2007.pdf
[5] L. Peterson, S. Shenker and J. Turner, Overcoming
the Internet Impasse through Virtualization, Hotnets 2004.
[6] J. Turner and D. Taylor, Diversifying the Internet, Proceedings of Globecom 2005.
[7] J. Touch, Y. Wang, L. Eggert, , and G. Finn, A
Virtual Internet Architecture, ACM SIGCOMM Workshop on Future
Directions in Network Architecture, 2003. Karlsruhe, Germany.
[8] Larry Peterson, Tom Anderson, Dan Blumenthal,
Dean Casey, David Clark, Deborah Estrin, Joe Evans, Dipankar
Raychaudhuri, Mike Reiter, Jennifer Rexford, Scott Shenker, and John
Wroclawski, "GENI design principles," in IEEE Computer, September 2006
[9] Clean slate project, http://cleanslate.stanford.edu [10]
ftp://ftp.cordis.europa.eu/pub/ist/docs/ct/internet-of-the-future-cluster_en.pdf
With kind regards,
Martin Stiemerling
NEC Laboratories Europe & University of Goettingen
http://www.nw.neclab.eu/ http://www.net.informatik.uni-goettingen.de/people/martin_stiemerling
------
stiemerling@nw.neclab.eu <== NEW ADDRESS
NEC Laboratories Europe - Network Research Division NEC Europe Limited
| Registered Office: NEC House, 1 Victoria Road, London W3 6BL |
Registered in England 2832014