Future Internet Research @ IETF75

From $1

Wednesday 29th July, 2009, 1145-1400 hrs, room B1, Clarion Hotel Sign, Stockholm


Introduction Mat Ford (ISOC)

'For Success of Future Internet Research' Dae Young Kim (CNU, Korea)

A Step Forward Jun Murai (WIDE, Keio University)

EIFFEL: Agreeing on Disagreements Dirk Trossen (BT)

Future Internet Xing Li (CERNET, Tsinghua Uni)

4WARD Henrik Abramowicz (Ericsson)

US Future Internet Research Joe Touch (USC-ISI)

Some reference material

EIFFEL White Paper, July 2009

Active Internet Measurements (AIMS) workshop report

NetSE Research Agenda, Version 1.0, July 2009

Key points of discussion

This meeting is part of ISOC's ambition to bring the community together to discuss cross-regional issues for Internet research and to forge an understanding of what the community thinks the priorities are. ISOC is also interested in understanding what they (ISOC) could do that would most benefit the research community in this area.

Co-ordination between different geographic regions occurs at funding agency level, but at the same time those funding agencies make it difficult to have projects that cross their region.

There are lots of venues to talk about FI research, what is lacking is coherent leadership to say, 'these are the problems that are important, these are short-term, these are long-term'.

Lucy Lynch observed that in her work on Trust & Identity the intention isn't so much to provide leadership as it is to provide incentive - to provide a taxonomy of existing initiatives, identifying where the gaps are, what succesful deployment would look like, and so on. The aim is to be engaged in the dialogue and to provide some context for it.

Ensuring the research remains connected with the real-world concerns and issues of the operational community is also a concern for some.

Joe Touch particularly emphasised the importance of the Science of Networking (not Networking Science), i.e. trying to get beyond Shannon and Information theory. Shannon says 'given a point-to-point channel between two parties that are trying to communicate...' Touch is interested in understanding how you find the party you wish to communicate with, how the intial peering relationship is formed. Touch's project on Recursive Network Architecture says, 'what if we still have protocol layers, but they all run essentially the same protocol?'. They have developed a mechanism to show how BGP, ARP, DNS, Google are all the same thing. IP forwarding and switching is the same thing as what ARP is doing. So instead of having top-down recursion we have tail recursion and this is starting to look like a science. We no longer break things into seven layers because it is easier to program that way, we break things into seven layers because there are seven layers fundamentally there. 'Structured Streams' is Bryan Ford's work which is along the same lines.

There was some interesting discussion of different approaches to problem solving and the idea that often long-standing problems were solved by those who removed the problem entirely by adopting fundamentally different approaches than had been attempted in the past.

It was also pointed out that above problem solving there exists another form of Internet research which is enabling new things to happen. One might think of information networking, pub-sub, as an example of that kind of research. Within the problem solving area, it's very different to take an engineering approach to a problem because something's new (like the Internet was when it first started - it was largely engineered by innovative thinkers), but once something is a global artefact, you need a balance between that approach, the 'fix it' type approach, and the scientific approach. You need models and more abstract understanding. This requires a different sort of community than that which traditionally attends the IETF and which started the Internet - this is a new era we're moving into.

One concern raised was that many different types of industry are now talking about use of IP technology, e.g. the automotive industry, the consumer electronics industry. The automotive industry is interested in standardising the use of IPv6 for in-car networks in ISO. There is a risk that different industries will choose different standards bodies - they'll all be using IP, but there'll be lots of divergence. Separate islands of IP technology are coming if we can't provide a common venue for discussion across these industries and SDOs.

The meeting exposed and highlighted the really wide range of different views with regard to Internet research priorities.

Guidance and opportunities for ISOC

There was disagreement over whether an ISOC-formulated research agenda would attract any real attention. There is a natural 'tussle' between funding agencies and researchers themselves over what the research agenda should be.
It was felt that leveraging colocation with IETF was very beneficial for improving the networking impact of such a meeting.

It was felt that there may be a role for ISOC in leading the way to specific research publication venues and helping draw attention to those to spotlight the best high-quality research from around the globe. One possibility suggested was to pick one or two research venues and summarise the output and issues that need attention to a wider audience.

It was suggested that ISOC might provide a central repository of venues like CoNext, ICCCN, Globecomm, SIGCOMM, Infocom, ICC, etc.

Using the network itself to help support the community (e.g. through web-based repositories) was highlighted as particularly beneficial at a time when restrictions on physical travel are widespread. Moving the debate online and providing a record of reports on state-of-the-art were areas that could help minimise duplication of effort and increase the efficiency of the global collaboration.

It was observed that in addition to the national/regional initiatives highlighted as part of this meeting, the ITU has also recently initiated a Focus Group on Future Internet, so having some kind of umbrella activity would be very nice to help counter the excess of meetings and venues for these discussions.

In conclusion, it seems clear there is likely a role for ISOC to highlight particularly important challenges for the Internet that the research community could focus on, and also to highlight the key venues for socialising research results. The IETF meeting itself is a good forum for cross-fertilisation of researcher ideas and for physical meetings and the IRTF is a key piece of that. ISOC is not looking to create venues where perfectly fine venues exist already, so our interests are in facilitating things that aren't better facilitated by others.


Name Affiliation
Mat Ford ISOC
Phil Roberts ISOC
Leslie Daigle ISOC
Alejandro Acosta ISOC Fellow, BT
Paul Hoffman VPN Consortium, PKNG RG
John Schnizlein ISOC
Dirk Trossen BT
Eduardo Reis NIC.BR
Jianping Wu CERNET, Tsinghua
Xia Yzn CERNET, Tsinghua
Marcelo Bagnulo UC3M
CS Hong Kyung Hee University
Gonazlo Camarillo Ericsson
Zhiking Wang CERNET, Tsinghua
Henrik Abramowicz Ericsson
Hannu Flinck Nokia Siemens Networks
Ove Strandberg Nokia Siemens Networks
Dimitri Papadimitriou Alcatel Lucent Bell Labs
Kwok Ho Chan Huawei
Tony Li Ericsson
Rolf Winter NEC
Ryuji Wakikawa ToyotaITC/WIDE
Jun Murai WIDE/Keio University
Aminul Chowdhury Ajou University
Eun Kyoung Paik Central R&D, KT
Haris Shamsi ISOC Fellow
Lars Eggert Nokia / TKK
Jorg Ott TKK Comnet
Kevin Fall Intel Labs Berkeley
Volker Hilt Alcatel Lucent Bell Labs
Rajeev Koodli Starent Networks
Christian Vogt Ericsson
Lucy Lynch ISOC
Joe Touch USC/ISI
Jun Bi CERNET, Tsinghua
Dan Massey Colorado State University
Lixia Zhang UCLA
Lynn St. Amour ISOC
Congxiao Bao CERNET
Jukka Manner TKK Comnet / Futureinternet.fi
Bryan Ford MPISWS / Yale University
Philip Eardley BT
Bob Briscoe BT
Hamid Mukhtar ETRI
Dae Young Kim CNU, Korea

FileSizeDateAttached by 
Jun Murai's presentation
4.99 MB15:53, 10 Aug 2009Ford@isoc.orgActions
 EIFFEL ISOC meeting Stockholm 290709-1.pdf
Dirk Trossen's presentation
766.29 kB15:53, 10 Aug 2009Ford@isoc.orgActions
EIFFEL White Paper
399.07 kB16:02, 10 Aug 2009Ford@isoc.orgActions
Introductory presentation
1984.53 kB15:56, 10 Aug 2009Ford@isoc.orgActions
Xing Li's presentation
696.61 kB15:56, 10 Aug 2009Ford@isoc.orgActions
Henrik Abramowicz's presentation
422.67 kB08:44, 11 Aug 2009Ford@isoc.orgActions
NetSE Research Agenda
1858.52 kB16:02, 10 Aug 2009Ford@isoc.orgActions
Active Internet Measurements (AIMS), workshop report
101.46 kB16:02, 10 Aug 2009Ford@isoc.orgActions
 ietf75 panel - dyk.pdf
Dae Young Kim's presentation
28.68 kB09:18, 12 Aug 2009Ford@isoc.orgActions
Joe Touch's presentation
1372.57 kB15:53, 10 Aug 2009Ford@isoc.orgActions
Images (0)
Comments (0)
You must login to post a comment.