GreenTouch Consortium: Building the Roadmap

1. GreenTouch Consortium:
Building the Roadmap
Dan Kilper
Chair, Technical Committee

2. 2020 ICT Carbon Footprint
820m tons CO2
360m tons CO2
2007 Worldwide ICT
carbon footprint:
2% = 830 m tons
CO2
Comparable to the
global aviation
industry
260m tons CO2
Expected to grow
The Climate Group, GeSI
to 4% by 2020 Report ―Smart 2020‖, 2008
© 2011-2012 GreenTouch Consortium

3. Tightening Energy Bottleneck
Bandwidth, Services, A
pplications
 Capacity limits for  Deploying new
existing technology and Energy efficient
by power/thermal equipment
budgets
 Publicly visible good-will
provided by being "Green"  With energy costs sky-
 Proactive stance in reducing Internet rocketing, increasing share
of operations cost
greenhouse gas emissions
 Cooling is high percentage
 Customers feel better about
of network power
service provider . . . leading
consumption (20% to 50%)
to revenue growth
© 2011-2012 GreenTouch Consortium

4. Fiber Capacity Deficit or “Crunch”
Data
Penny Calls, What next?
24/7 Connection
10x
Gb/s
Traffic
Voice
Only call on Commercial
weekends Systems
1980 1990 2000 2010 2020
Year
© 2011-2012 GreenTouch Consortium

5. The Network Energy Gap
50
40 Mobile Traffic Internet
Data Backbone
30
Growth
Growing Gap!
20 Mobile
Efficiency
10
Wireline
Efficiency
2005 2010 2015 2020
Year Kilper, et. al., IEEE JSTQE 2011
© 2011-2012 GreenTouch Consortium

6. Ideal Efficiency Improvements on BAU: Flat in 2020
 Improving network efficiency at best keeps power consumption flat
over next decade
 What happens after 2020?
 Can only use ‗sleep modes‘ once
100 BAU Current technology will only
Power/User (W)
sustain us for another decade:
how do we go beyond?
10
Optimistic
Improvements
1
2008 2010 2012 2014 2016 2018 2020
Year
© 2011-2012 GreenTouch Consortium

7. ICT Industry Responds
 First Step: metrics, awareness, standards, call to action
 Global e-Sustainability Initiative (GeSI)
 ITU-T
 GreenGrid
Semiconductor Industry
Already Organized for the
 Next Step: Cooperation, Action Long View:
 EARTH: LTE 2x ITRS 15 years out
 Mobile VCE: Green Radio 100x
 Institute for Energy Efficiency: Wireless and Optical 100x
 GreenTouch: ICT Networks 1000x
© 2011-2012 GreenTouch Consortium

8. What is GreenTouch?
GreenTouch Mission
By 2015, our goal is to deliver the architecture, specifications and
roadmap — and demonstrate key components — needed to increase
network energy efficiency by a factor of 1000 from current levels.
 Broad, open and global consortium executing research projects to
achieve aggressive goal
 Roadmap organization establishing reference architectures and
research targets to overcome major challenges facing network
scaling and energy
 Venue for cooperation and enabling demonstrations among research
organizations
 Forum for the exchange of information on energy trends,
challenges, & research on communication networks
© 2011-2012 GreenTouch Consortium

9. GreenTouch Members
 AT&T Services  France Telecom  Swisscom
 Athens Information  Freescale Semiconductor  TNO
Technology (AIT) Center  Fujitsu  Tsinghua University
for Research & Education  Huawei  TTI
 Bell Labs, Alcatel-Lucent  IBBT  TU Dresden
 Broadcom  IMEC  University College London
 Carnegie Mellon  INRIA  University of Cambridge
University
 CEA-LETI Applied  KAIST  University of Delaware
Research Institute for  Karlsruhe Institute of  University of L‘Aquila
Microelectronics Technology  University of Leeds
 China Mobile  Katholieke Universiteit  University of Manchester
 Chunghwa Telecom Leuven (K.U. Leuven)  University of Maryland
 Columbia University  King Abdulaziz City for  University of Melbourne‘s
Science and Technology Institute for a Broadband-
 Commscope/Andrew  KT Corporation
 Draka Communications Enabled Society (IBES)
 National ICTA Australia  University of New South
 Dublin City University  Nippon Telegraph and Wales
 ETRI Telephone Corp  University of Paderborn
 ES Network/Lawrence  Politecnico di Torino  University of Rochester
Berkeley Labs  Portugal Telecom
 Fondazione Politecnico di  University of Toronto
Inovação, S.A.  Waterford Institute of
Milano  Samsung (SAIT)
 Fraunhofer-Geselleschaft Technology
 Seoul National University  ZTE
GreenTouch : Building the Roadmap | 2011
© 2011-2012 GreenTouch Consortium

10. GreenTouch Organization
Executive Board
Gee Rittenhouse
Technical Committee Operations Committee
Dan Kilper, Shugong Xu Thierry V. Landegem, Kevin Kemp
Services, Applications Trends Network Committee
Steve Korotky , Bian Sen Kerry Hinton
Working
Groups
© 2011-2012 GreenTouch Consortium

11. GreenTouch Working Groups
Working
Groups
Access Networks
Wireline Access Mobile Communications Core Networks
Networks U. Barth,
P. Vetter, L. Lefevre E. Calvanese-Strinati
Optical Networking & Switching
Transmission and Routing
W. Shieh, C. Dorize T. Klein, J. Elmirghani
© 2011-2012 GreenTouch Consortium

12. GreenTouch Approach
 Bottom Up Research Organization
 Use of models to structure and guide research and
collaboration
 Funding through member contributions & external
sources
 Gauge impact of innovations on:
 Alternative metrics (carbon footprint, network power,
embedded energy)
 Adjacent technologies (data centers, handsets)
 Measure, model and predict energy consumption in ICT
networks (equipment trends, traffic, deployment)
© 2011-2012 GreenTouch Consortium

13. Targets and Challenges
2
How do we
10 prevent
this?
What are
Power/User (W)
1
10 Wireline Access
the major
technical
10
0
obstacles?
If we wait
-1
10 until
2020, won‘
t see
2010 2015 2020 solution
until 2030!
Year
© 2011-2012 GreenTouch Consortium

14. Consortium 5 Year Goal
100  Define architectures
1000x Target
 Demonstrate tech.
10
Use models for
Efficiency (Mb/s/W)
network in 2020 to
1
0.1 set technology
demonstration
0.01 Total Network: requirements
BAU
1E-3
1E-4
2010 2015 2020
© 2011-2012 GreenTouch Consortium
Year

15. Use Architecture Models &
Targets to Track Progress
1000
Goal
Network Eff. Improvement
Remaining
targets achieved
100
12 targets
achieved
1 target
10 achieved
3 targets Architecture 1
Architecture 2
achieved Architecture 3
1
2010 2011 2012 2013 2014 2015 2016
Year
 Define architectures and track research results
 Identify targets for each architecture and update network efficiency in model
as targets are achieved
• Working groups define targets and evaluate completion
• Targets can be achieved within GreenTouch projects or from broader community
• Identify gaps in effort and solicit new activities
© 2011-2012 GreenTouch Consortium

16. Modeling Provides Clear Picture of Goals
• Baseline year 2010
• Target year 2020
• GreenTouch results in five years: 2015
DRAFT
1.E+04
Efficiency in Mbps/W
GreenTouch 5 year Goal: Element efficiency demonstration
MODEL
targets for model 2020 network
1.E+03 Overall network efficiency
target in 2020
Efficiency in Target Year
1.E+02
1.E+01
1.E+00
1.E-01
1.E-02
1.E-03
1.E-04
Access Transport Rtg&Swg Mobile Overall
© 2011-2012 GreenTouch Consortium

17. Roadmap Details Underway
GPON XGPON
After sleep mode  Draft wireline access
2.0
cpe trend
Efficient HW design  Draft Core Opt Ntwk
1.8
transponder model
1.6
Virtual HGW
1.4 Tx: P(W) Notes
BI PON Mux 0.40 10pJ/b
1.2 Driver 1.00 25pJ/b
(RF overlay)
Home gateway processor Laser 0.06 1.5pJ/b
Watt/User
Voice user interface Mod 0.16 4pJ/b
1.0
Data user interface (Ethernet and Wireless) Subtotal 1.62
PON digital
0.8 OE PON Rx:
PhotoRx 0.144 3.6pJ/b
0.6 Low power circuits DeMux
Subtotal
0.80
0.95
20pJ/b
beyond CMOS
0.4 2.56W 20%
+Inefficiencie 10.25W 80%
s
0.2
Pfunction 12.81W
Overheads 12.81W
0.0
Total 25.6W
2010 2011 2012 2013 2016 2017 2020
© 2011-2012 GreenTouch Consortium

18. Structure
Services, Applications & Trends
Mobile Wireline Access Core Switching Core Optical
Communications Networks & Routing Networking &
Transmission
2000X 1600X 400X 20X
© 2011-2012 GreenTouch Consortium

19. Some Research Projects…
Beyond Cellular – Green Mobile Networks
Virtual Home Gateway
Optimal End-to-End Resource Allocation
Service Energy Aware Optical Networks Steady
Green Transmission Technologies flow of
new
Minimum Energy Access Architectures
Projects
Single-Chip Linecards & coop.
Large-Scale Antenna Systems activities
Highly-Adaptive Layer Mesh Networks
Massive MIMO
© 2011-2012 GreenTouch Consortium

20. First Technology Demonstration: Large Scale Antenna
Array Systems, Using MIMO to Focus RF Energy
Processing
Unit
Marzetta, T. L., IEEE Trans
Wireless Comm, Nov 2010
© 2011-2012 GreenTouch Consortium

21. Current Antenna Technology
Antenna
End-user
© 2011-2012 GreenTouch Consortium

22. Current Antenna Technology
Antenna
End-user
Power used
© 2011-2012 GreenTouch ConsortiumConsortium
© 2011 GreenTouch

23. Simulating Current Antenna Technology
16 element antenna array
Antenna
End-user
Power used = 16W
© 2011-2012 GreenTouch Consortium

24. Large Scale Antenna System Demonstration
Antenna
16 Antennas  16x power reduction!
Collaborators:
Bell Labs
Freescale
Huawei
imec End-user
Samsung
Power used = 1W
© 2011-2012 GreenTouch Consortium

25. Beyond Cellular Mobile
Separating Data Network from Signaling Network
Data Network Signaling Network
Collaborators:
Polimi
Bell Labs
Huawei
Samsung
INRIA
Mobile Communications Working Group
© 2011-2012 GreenTouch Consortium

26. Wireline Access DSP Platforms
for Wireless
Min. Energy Access Sleep Modes Home Network
Architectures Un-cooled Tunable
Lasers
Virtual Home Low Power Novel PON
Gateway OFDM in Optical Protocols; Low Fiber in the
Access Power Customer Home
Premises Equipment
© 2011-2012 GreenTouch Consortium

27. Single-Chip Router Linecards
Multiple Cores
Cross-X
Optical
Fiber
Silicon CMOS
Photonics
Content
Cache
Hybrid Electronic – Photonic Chip
Core Switching & Routing Working Group
© 2011-2012 GreenTouch Consortium

28. Cooperation Among Diverse Research Groups
& Consortia
 Cooperating Project INTERNET: INTelligent Energy awaRe NETworks
Input
Thru port
Drop port
Add port
Backplane
Line cards
Zinwave
© 2011-2012 GreenTouch Consortium

29. Current GreenTouch Projects
 BCG2: Beyond Cellular Green Generation*
 GTT: Green Transmission Technologies* *Cluster project made up of
 LSAS: Large Scale Antenna Systems* several sub-projects/activities
 Minimum Energy Access Architectures
 Minimum energy access architecture demonstration
 VHG: Virtual Home Gateway
 OPERA: Optimal End to End Resource Allocation
 STAR: Switching & Transmission
 REPTILE: Router Power Measurements
 Single Chip Linecard
 ZeBRA: Zero Buffer Router Architectures
 SEASON: Service Energy Aware Sustainable Optical Networks*
 HALF MOON: Highly Adaptive Layer for Mesh On-off Optical Networks
 EFICOST: Energy Efficient High Capacity OFDM Signal Transmission
 Telecommunication Audits and Data Aggregation
© 2011-2012 GreenTouch Consortium

30. GreenTouch Project Process
Project concept
Technical Committee
presented to
discusses and
relevant WG(s); Relevant WG(s)
approves the
project proposal discusses and
proposal; passes
prepared endorses the
funding requests
incorporating any in- proposal
onto the Executive
kind or cash funding
Board
from partners
Includes endorsement
of contributed or
cooperative projects
& activities
Opportunity for Members can direct
members to donations for the purpose
review, solicit, & fund of funding projects or
projects or make in-kind activities, or make a
contributions general contribution
© 2011-2012 GreenTouch Consortium

31. Initial Activities
 Build research program
 First technology demonstration spring 2011
 Establish common reference architecture
 Define primary research targets
 Establish expected trends on key metrics to 2020
 Provide international forum for cooperation and
exchange of ideas on energy research topics
© 2011-2012 GreenTouch Consortium

32. New Approaches: Focus on Energy
 New devices
• Analog vs digital, best use of optics and electronics
• Old ideas finding new life: large scale MIMO
 New architectures
• trade-off transmission/bandwidth and processing, distributed versus
centralized
 New protocols
• Longer packet sizes or no packets at all for certain applications
 Service optimized networks
• Move away from one size fits all—use most energy efficiency hardware for
the service
• Coordinate service delivery/applications with network hardware operation
 Restructuring layers, architectures, feature options
• How much do way pay in energy for convenience? duplicated functions (FEC)?
• What technologies do we really need in order to support the essential
capabilities?
© 2011-2012 GreenTouch Consortium

33. Minimizing Processing
 Repetition
• Unnecessary router hops
• Inter-operator exchange Separate what is
• Multiple transmissions needed from what
 Remove processing from the data path is convenience
• Separate control channel?
 Focus on Service
• Content delivery vs. browsing vs messaging
 Push to the edges
• FEC
• Security, policy processing
 Simplified Addressing
• Geographic addressing/binary switching
© 2011-2012 GreenTouch Consortium

34. A Scalable Internet: Holistic Re-Design
 Processing
• New addressing—transparent data flow
• End-to-End—security, FEC
 Back to the Future
• Optimized hardware for given task: service differentiation
– Take the movies out of the routers
 How many layers do we need?
 How do protocols and algorithms impact hardware design?
 How do we unlock physical potential—small cells, efficient MIMO?
 What is the real energy cost for the features and functions
supported in the network?
© 2011-2012 GreenTouch Consortium

35. Opportunities
 A global organization dedicated to the long
300 term sustainability of ICT networks
ANNUAL GROWTH RATE (%)
 Tackle major problems end-to-end, full
250
picture
200  Lay the foundation technologies
150  Clear vision for future: roadmap
100 Future is a different
game than the past
50
0
1995 2000 2005 2010 2015 2020 2025
YEAR
© 2011-2012 GreenTouch Consortium

36. Thank You
www.greentouch.org
© 2011 GreenTouch Consortium