Future Internet Research
and Experiments
EU vs USA
7 June, 2012
Róbert Szabó
Dept. of Telecommunications and Media Informatics
Budapest University of Technology and Economics
Agenda
My personal interpretation of the
Open Platforms for Innovation session
at FIA Aalborg
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
2
Programs
EU
 FI – PPP
USA
 GENI
 IGNITE
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
3
FI – PPP (EU)
 Industry driven initiative set to build an open Future Internet
network and service platform
 Platform instances validation in large scale real-life trials
 Broad-based collection of Future Internet technical
requirements
 Identification of legal, regulatory and standardization
requirements
 Modeling novel business models and value networks
 New methods and tools to engage SMEs and foster
entrepreneurship
 Explorative funding instrument to do research create
communities
Source: Dr. Petra Turkama, Aalto University, CONCORD CSA
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
4
GENI – Infrastructure (USA)
“Exploring Networks of the Future
The start of GENI campus expansion”
GENI is a virtual laboratory for
exploring future internets at scale
GENI opens up huge new opportunities
 Leading-edge research in next-generation
internets
 Rapid innovation in novel, large-scale
applications
Source: Chip Elliott, GENI Project Director
EU vs US
Universities and Campuses
Vs.
Source: Dr. Petra Turkama, Aalto University, CONCORD CSA
Participant Type (by EC funding)
7%
2%
20%
Res earch organis ations
3%
Public body
SME
8%
Private com m ercial
HES
Other
60%
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
6
How to Sponsor the Industry?
(incentives)
EU
 Direct involvement
USA
 By orders from the programs
 The research community decides what to
use  what to order
• E.g., Open Flow enabled switches and routers
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
7
GENI’s growing footprint
Source: Chip Elliott, GENI Project Director
GENI architecture
U N I
ISP
V
E
R S
I
T
Y
Internet
Metro
Research
Backbones
g
GENI-enabled
hardware
Legend
Layer 2
Data Plane
Layer 3
Control Plane
g
U N I V E R S I T Y
Regional Networks
Campus
g
 Flexible network / cloud
 Support “hybrid circuit” model
research infrastructure
plus much more (OpenFlow)
 Also suitable for physics,
 Distributed cloud (racks) for
genomics, other domain science
content caching, acceleration,
etc.
Source: Chip Elliott, GENI Project Director
GENI “at scale”
Suggest 100-200 US campuses as
target for “at scale”
 Both academia and national labs
 GENI-enable the campuses
 Their students, faculty, staff can then
“live in the future” using both today’s
Internet and many experiments
 Build out backbones, regionals, and shared
clouds to support the campuses
Source: Chip Elliott, GENI Project Director
Enabling “at scale” experiments
 How can we afford / build GENI at sufficient scale?
 Clearly infeasible to build research testbed “as big as the Internet”
 Therefore we are “GENI-enabling” testbeds, commercial equipment,
campuses, regional and backbone networks
 Students are early adopters / participants in at-scale experiments
 Key strategy for building an at-scale suite of infrastructure
HP ProCurve 5400 Switch
NEC WiMAX Base Station
GENI-enabled
equipment
GENI-enabled campuses,
“At scale” GENI prototype
students as early adopters
Source: Chip Elliott, GENI Project Director
US IGNITE
• Science: Advances in networking and systems research will be
enabled through experimentation and explorations at scale on
unique network infrastructure.
• Innovation: Leveraging previous NSF-investments in the GENI
network will allow use of an open testbed available across the
globe for experiments in future networks; will enable
deployment and evaluation of gigabit applications that were
previously not possible.
• Societal Benefit: Applications in national priority areas (e.g.,
health and wellbeing, emergency preparedness, advanced
manufacturing, energy, education and workforce development,
transportation) may have profound, long-term social and
economic impacts.
Source: Chip Elliott, GENI Project Director
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
12
US Ignite is now taking shape
Bridging CS Experiments to Next-Gen Applications in Cities
US Ignite Initiative
GENI
Research Infrastructure
for Computer Scientists
Testbed for
Next-Gen Applications
Future commercial
offerings
federation
GENI technology
GENI members, policies, …
US Ignite participants, policies, …
Campus and Lab
Applied Research
CS Research
CS Experiments
App creation teams
Pre-commercial Applications
Experimental Usage and Demonstrations
Campus networks
Service creators
Municipal and
commercial networks
Regional and backbone networks
Commercial Applications
IGNITE Arch Proposal
Existing head-end
Existing
ISP
connects
Most equipment
not shown
Layer 3 GENI
control plane
Layer 2
Ignite
Connect
(1 GE or
10GE)
Layer 2 connect to
subscribers
OpenFlow switch(es)
Flowvisor
Remote management
Instrumentation
Aggregate manager
Measurement
Programmable PCs
Storage
Video switch (opt)
New GENI / Ignite rack pair
Home
FI PPP First Year Achivements
 Negotiation and implementation of FI PPP Collaboration
Agreement and Governance structures
 Selection of Advisory Board
 Release the first set of requirements
 Release the first platform instances
 Cross-project collaborations in Working Groups and
workshops
 Identification of potential test beds for experimentation
 Promotion of FI PPP in various forums and communities
 Building links and finding synergies with related initiatives
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
15
Conclusions
Support research at Universities and
Research Labs
Support experiments at campuses
Go large scale and open
Let the industry find profitable
business in the above
Robert Szabo (BME-TMIT)
<robert.szabo@tmit.bme.hu>
16