FIND: Future Internet Design

Report
FIND: Future Internet Design
Informational Meeting
November 7, 2006
Darleen Fisher
CISE
National Science Foundation
[email protected]
FIND Challenge to the Research
Community:
Create the Future Internet you
want to have in 10-15 years
The Future Internet
Must
• Be worthy of our society’s trust
– Even for managing and operating critical infrastructures
• Provide a bridge between physical and virtual worlds
– Via instrumented and managed sensorized physical environment
• Support pervasive computing
– From wireless devices to supercomputers
– From wireless channels to all optical light-paths
• Enable further innovations in S&E research
– Seamless access to networked instruments, supercomputers,
storage, etc.
• Create a social world in which we would want to live
3
What is Different This Time?
• Clean-slate approach
– To overcome Internet ossification
– Research not constrained by the features of the
current Internet
– But does not mandate rejecting what currently
works
• A comprehensive coordinated effort
– Ability to try different approaches (We do not
have a preconceived idea of what they are)
• Ability to experiment at scale
– With real users and applications
4
Success Scenarios
• Internet evolution influenced by clean-slate approach
• Alternate Internet architecture emerges
– Alternate architecture(s) coexist with the current Internet
– Virtualization becomes the norm with plurality of architectures
– Single architecture emerges and dominates
• New services and applications enabled
• Many other payoffs
– Some unexpected
5
FIND - Different Process
•
•
•
•
•
“Goal” oriented  Future Internet
– Not typical for NSF research programs
– Area has a longer timescale with sustained funding
Three phases -- iterative and overlapping
– Exploration of architectural components and 1st cut overarching
architectures
– Convergence into multiple full-scale architectures
– Experimentation of architectures at scale
“Competitive cooperation” model
– Competition to bring out the best
– Cooperation to build on each others work to deliver Future
Internet
Competition – Proposal reviews
Cooperation – Among awardees
– Regular meetings -- three times a year
– Commitment to openness and transparency
6
Stages of Research
2006-2007
• Research on architectural elements
– Naming, identities, forwarding, interdomain
protocols, etc.
• 1st Cut overarching architectures
• Cross-cutting requirements of built-in
security, robustness, and manageability etc.
• Transformational architectural ideas (like
packet switching was for the Internet)
7
Stages of Research
2007 and Later
Coordinated effort to assemble overarching
coherent architectures
• Multiple PI meetings to formulate architectures
– FIND awardees
– But also other “architectural” researchers e.g. funded
by NeTS, CyberTrust, DARPA, industry, internationally
funded researchers, etc.
• Small number of architectures developed
8
Stages of Research
2008 and Later
Architectures as they emerge will be made
operational and tested
– Simulation
– Emulation
– Run on a large-scale GENI facility
• Experiment with new architectures at scale
9
Where are we now?
10
FIND 2006
26 of 98 projects awarded
– some 1 year seed investments
NeTS = $40M
FIND = ~$15M (38% of NeTS)
11
FIND 2007
• NSF 07-507; January 22, 2007 deadline
• First FIND PI meetings
– Begin to create a FIND community and identify areas
of commonality and differences, open research areas
• FIND project descriptions at:
http://nsf-find.cs.umn.edu
• Process for including FIND-like researchers from
industry, international and academics funded
elsewhere
12
FIND Portfolio
• New architecture principles
– Composable architectural building blocks
– Recursive network architecture
• Delay tolerant network architectures
– Disaster networks
– Cache and forward network (for large files)
• Network technology and architectures
– Wireless Networks
– Optical Networks
• Higher bar—be more architectural
• Services architectures
13
FIND Portfolio
• Virtualization
– New “layer 2+” hour glass
– New e2e services from virtual nodes & links
(multiple hour glasses)
• Sensor networks
• Security
– Default off and least knowledge
– Packet attribution with privacy preservation
• Routing
– User controlled routes
14
FIND Portfolio
• Object identity and naming
– Handles for naming, security &
communication paradigms
– Usable namespaces for small devices
• Economics and network architectures
– Market enabling architectures
• Management
– Manageability in routing systems
– Model-based diagnosis of the knowledge
plane
15
Lessons from FIND 2006
• Is it networking? Does it belong in NeTS?
• Is it architectural? Does it belong in FIND? How well
does it address architectural issues?
• Is it “clean slate”? Is it appropriate for a “Future
Network”?
• Is there evidence of a deep understanding of the
(hard) issues?
• Will it scale?
• Can it be secure?
• What about overhead or complexity?
• Is it economically viable?
• Have ideas already been tried?—need good
understanding of the literature and history of the
field when rethinking old ideas
16
Lessons from FIND 2006
• Is there a unifying theme of ideas—or just a
“grab bag” of ideas?
• Is there an evaluation plan for the new ideas?
• Is there a well developed plan for the integration
of education and research?
• How will the PIs involve students in the FIND
research?
• What is the broader impact if successful? Does
the idea matter? Want to hear “if successful,
these ideas would have a profound positive
impact on the Future Network.”
17
Lessons from FIND 2006
• Too many ideas merely extensions of
current Internet or P2P networks
• Too many were technology, not
architecture proposals
• Too many use assumptions appropriate for
the current Internet
• Not just “fix the Internet” mentality
• It does NOT have to fit within today’s
vision of GENI’s capability—GENI will
evolve
18
Guidance to FIND Proposers
• Think of creative yet well-considered ideas
• Clearly show how the proposed work
addresses one or more Future Internet
requirements (e.g. built-in security,
economic viability, manageability)
• May submit architectural components,
architectural theory, 1st cut overarching
Architectures (not just critique of Internet)
• Discuss how your work would fit into a
larger overall network architecture
19
FIND Review Criteria
• Intellectual Merit
• Broader Impact
• How well the work addresses architectural
requirements
• How well the proposed work relates to and
enhances overall architectural framework
• Importance of work to framing a new
architecture
20
NeTS Program Directors 2007
• FIND—Darleen Fisher & Allison Mankin
• NBD—Darleen Fisher
• WN—David Goodman (leaves 2/06)
– Recruiting new PD with wireless networking
expertise
• NOSS—David Du
21
Allison Mankin
• Co-program director FIND (with Darleen Fisher)
• Co-program director GENI (with Guru Parulkar)
– Consultant, Shinkuro, Inc., Bell Labs, USC/ISI, NRL, U. Wisc
(visiting scientist), MITRE
– Author of many published networking research papers
– Co-editor IPng: Internet Protocol Next Generation 1995
– Co-Director, IETF Process for Selection of the Next Generation
Internet Protocol
– Director, CAIRN (successor to DARTnet)
– Internet2 Abilene Technical AC
– Area Director, Internet Engineering Steering Group
– Chair, IETF Geolocation Privacy WG (ongoing)
– ICANN Security & Stability Committee
– Member of various boards, directorates, and working groups
22
David Clark
“FIND Architecture and Outreach Coordinator”
• Remain a member of the research community,
but work with NSF & community
– Plan PI meetings
– Identify FIND research priorities
– Lead FIND team-building
– Help FIND researchers frame new architectures
– Outreach to researchers funded elsewhere
– Outreach to international FIND-like researchers
– Provide linkage between FIND and GENI
23
9:00-10:15
Welcome (Grand Dominion I-IV)
Darleen Fisher, Program Director, NeTS
FIND Program Past and Future
Introduction of Allison Mankin, Program Director, NeTS
Introduction of David Clark, MIT, FIND Architecture and Outreach
Coordinator
The Challenge of Thinking Architecturally -- David Clark
10:15
Break (Upper Rotunda)
10:30-12:00
FIND Principal Investigator Panel-Experience in
Writing
Funded FIND Proposals (Grand Dominion I-IV)
Nick McKeown, Stanford
Nick Feamster, GA Tech
Ken Calvert, University of Kentucky
12:00-1:00 Lunch (Fairfax Dining Room 13)
1:00-2:00
FIND Reviewer Panel—Advice about Writing FIND
Proposals (Grand Dominion I-IV)
Jorg Liebeherr, University of Toronto
Craig Partridge, BBN
K. K. Ramakrishnan, AT&T Labs-Research
2:00-3:00
Questions and Answers
24

similar documents