Green Telecom & IT Workshop: Ee routing and networking thierry klein
1 like842 views
This document summarizes the work of the Core Switching and Routing Working Group, which focuses on improving the energy efficiency of network equipment and traffic routing. It notes that internet traffic is growing exponentially but technology improvements are slowing, resulting in a widening "energy gap". The group is researching more efficient hardware components, network architectures, traffic engineering techniques, and cross-layer optimizations. Its goal is to develop new technologies and algorithms to significantly improve network power efficiency over the next decade as traffic volumes continue rising rapidly.
![SOME SPECIFIC ROUTER LIMITATIONS
INTERCONNECTS Buffer Switch
Receive
Optics Framer
L2
Buffering
Fwd
Engine
Input
Queuing
Fabric
Interface
Sw itch
Fabric ENERGY DOES NOT
L1+L2 L3
Sw itch
Fabric
FOLLOW LOAD
L2 Output Sw itch Route r T1600 (640Gb/s )
Optics Framer Output Fabric
Buffering Fw d Fabric
Queuing Interface
Engine 7,000
[From Kharitonov 2009]
6,000
18 Chip-to-chip Interconnects Buffer Mem
5,000
PACKET SIZE IPv4 Cumulative 4,000
W
3,000
2,000
Route r T1600 (640Gb/s)
1,000
0
0% 20% 40% 60% 80% 100%
Load
D.Kharitonov, “Time-Domain Approach to
Energy Efficiency: High-Performance
Network Element Design” 2009 IEEE
GLOBECOM Workshops
http://www.caida.org/research/traffic-
analysis/pkt_size_distribution/graphs.xml](https://image.slidesharecdn.com/eeroutingandnetworking-thierryklein-120410231045-phpapp02/85/Green-Telecom-IT-Workshop-Ee-routing-and-networking-thierry-klein-5-320.jpg)
Green Telecom & IT Workshop: Ee routing and networking thierry klein
- 1. Energy Efficient Routing and Networking Dr. Thierry E. Klein Director, Bell Labs Research / Alcatel-Lucent Chair, Core Switching and Routing Working Group, GreenTouch
- 2. OVERVIEW • Core Switching and Routing Working Group • Technology Limitations • Energy Efficiency Challenges • Focus Statement • Membership • Energy Efficiency Improvement Opportunity • Research Targets • Key Research Projects and Activities
- 3. PAST AND ANTICIPATED INTERNET GROWTH ANNUAL GROWTH RATE (%) 300 250 Internet Traffic Growth Rate 200 RHK - NA 150 McKinsey - NA MINTS - Global Arbor - Global 100 50 24-53%/year 0 1995 2000 2005 2010 2015 2020 2025 SKK, 2010 (Sources: RHK, 2004; McKinsey, JPMorgan, AT&T, 2001; MINTS, 2009; Arbor, 2009). YEAR Courtesy of Steve Korotky
- 4. TRAFFIC GROWTH AND TECHNOLOGY SLOW- DOWN Traffic doubling every 2 years • 40% per year • 30x in 10 years • 1000x in 20 years Slow-down in technology • Network energy efficiency increasing 10-15% per year Leading to an energy gap 10000 CMOS Feature size (π) x0.88/yr 1000 Energy per bit Routed (nJ) 100 10 -1 IO (π ) -3 Logic (π ) 1 0.1 0.01 -4 Constant Router Power (π ) 0.001 1985 1990 1995 2000 2005 2010 2015 2020 Year
- 5. SOME SPECIFIC ROUTER LIMITATIONS INTERCONNECTS Buffer Switch Receive Optics Framer L2 Buffering Fwd Engine Input Queuing Fabric Interface Sw itch Fabric ENERGY DOES NOT L1+L2 L3 Sw itch Fabric FOLLOW LOAD L2 Output Sw itch Route r T1600 (640Gb/s ) Optics Framer Output Fabric Buffering Fw d Fabric Queuing Interface Engine 7,000 [From Kharitonov 2009] 6,000 18 Chip-to-chip Interconnects Buffer Mem 5,000 PACKET SIZE IPv4 Cumulative 4,000 W 3,000 2,000 Route r T1600 (640Gb/s) 1,000 0 0% 20% 40% 60% 80% 100% Load D.Kharitonov, “Time-Domain Approach to Energy Efficiency: High-Performance Network Element Design” 2009 IEEE GLOBECOM Workshops http://www.caida.org/research/traffic- analysis/pkt_size_distribution/graphs.xml
- 6. CORE SWITCHING AND ROUTING WORKING GROUP Focused on components, technologies, systems, algorithms and protocols at the data link layer (L2), the network layer (L3) and the transport layer (L4) as well as interactions with lower and higher layers and research efficiencies that can be obtained from cross-layer optimizations and joint designs • Network equipment hardware (routers and • Traffic engineering switches) • Bandwidth allocation & traffic grooming • Architecture and components • Elimination of over-provisioning • Functions, features and dimensioning • Efficient protection and restoration • Low energy technologies (including electronics, photonics, etc) • Multicasting, elimination of junk and redundant traffic …. • Power measurements • Network topologies and architectures • Network management, operation and • Tradeoff between optical and electronic data control transport Quality of service support • Optimal joint IP-optical network design Network-wide reconfiguration and control of • Packet versus circuit-switched architectures network elements (offline or online). Holistic, • Energy efficient and simplified routing end to end approach Protocols and algorithms for managing and • Integration of application and transport controlling network elements layers Control and data plane Cross-layer optimization for efficient content distribution Energy and traffic monitoring
- 7. WORKING GROUP MEMBERSHIP Athens Information Technology KAIST (AIT) Karlsruhe Institute of Technology Bell Labs (Chair: Thierry Klein) Nippon Telegraph and Telephone Broadcom Corporation Chunghwa Telecom Politecnico di Torino Columbia University Samsung Advanced Institute of Technology (SAIT) Dublin City University Seoul National University Electronics and Telecommunication Research University of Manchester Institute (ETRI) University of Melbourne Energy Sciences Network / University College London Lawrence Berkeley Labs University of Cambridge Politecnico di Milano University of Leeds (Co-Chair: Freescale Semiconductor Jaafar Elmirghani) Fujitsu University of New South Wales Huawei Technologies University of Toronto IBBT IIT Delhi 28 members organizations with 67 individual members INRIA
- 8. ENERGY EFFICIENCY IMPROVEMENT OPPORTUNITY • Provide assessment of potential opportunities for power efficiency improvements in packet networks • Include the electronically switched portion of a service provider network, including IP, Ethernet and OTN • Excluding fixed and wireless access networks • Excluding optical transport • Excluding opportunities for traffic reduction, e.g. via caching • Goal is to assess the opportunity for energy efficiency improvement with a realistic path towards realization within the GreenTouch timeframe • Determine “independent” dimensions so that power efficiency numbers can be multiplied to arrive at overall efficiency opportunity
- 9. BACKGROUND AND ASSUMPTIONS • Timeframe: • Consistent with GreenTouch timeframe • Algorithms, architectures and technologies that can be demonstrated by 2015 • With evolutionary improvements through 2020 • Applied to 2020 traffic • Comparison with 2010 traffic and 2010 technologies • Assumptions: • Consider the traditional IP packet data network framework • Alternative paths and technologies are possible, but more speculative and are not expected to fit in the timeframe: • Optical burst switching • Content centric networking • Adiabatic switching, quantum dot cellular automata, ….
- 10. OVERALL EFFICIENCY OPPORTUNITY • Defined 5 independent categories: • Chip level components and devices: 15x • Network element design: 1.5x • Network architecture: 2x • Dynamic resource management: 3x • Power utilization efficiency: 2x • Overall power efficiency opportunity: 270x • Caveats: • Numbers are best current estimates of efficiency improvement opportunity • Large degree of uncertainty especially around network element architecture and network architecture • Optimistic estimates since not clear if and how all the targets can be achieved • Pessimistic estimates since constrained to current IP packet network architectures and further-out technologies not considered
- 11. RESEARCH TARGETS BY FUNCTIONAL TOPIC (1) Research Target Target Chip Level Components and Devices Low power electronics and photonics 3x – 10x Opto-electronic integrated circuits 3x – 10x Network Element Design Scalable and energy efficient router architectures for peta-bit routers 1.5x Simplified and energy efficient protocols to eliminate unnecessary and redundant packet processing. Energy efficient software Integrated transceiver and wavelength circuit switching fabric operating in a core network to eliminate 10x routing infrastructure and reduce layer-2 switch energy/bit for targeted services Network Architecture Network architectures, topologies and joint IP-optical design 3x – 10x Energy efficient content routing (content router design, protocols and content placement and replacement 10x algorithms) Energy optimized combined source and channel coding designed for end to end service dependent efficiency
- 12. RESEARCH TARGET BY FUNCTIONAL TOPIC (2) Research Target Target Dynamic Resource Management Rate adaptation and sleep cycles (processors, buffers, switch fabrics, linecards, router) 2x – 4x Energy efficient routing 2x Energy aware scheduling algorithms designed for delay tolerant services that enable end to end buffereless transmission respecting service QoS requirements Power aware protection and restoration 2x Power Utilization Efficiency Passive cooling and advanced thermal management 1.5x • Requires equipment and network models with energy equations to determine overall energy efficiency improvement opportunity • Gain understanding into which research targets are additive and multiplicative • Gain understanding into most impactful research areas
- 13. SCORPION: SILICON PHOTONIC INTERCONNECT AND SINGLE CHIP LINECARD Contributing Members
- 14. OPERA: OPTIMAL END TO END RESOURCE ALLOCATION Contributing Members
- 15. STAR: SWITCHING AND TRANSMISSION 4x4 4x4 4x4 switch switch switch element element element 4x4 4x4 4x4 switch switch switch element Shuffle element Shuffle element 4x4 network 4x4 network 4x4 switch switch switch element element element Input Gating Output 4x4 4x4 4x4 shuffle SOAs shuffle switch switch switch network network element element element IP layer WDM layer Contributing Members
- 16. ZEBRA: ZERO BUFFER ROUTER ARCHITECTURE Contributing Members
- 17. REPTILE: ROUTER POWER MEASUREMENTS MODULAR SERVICES CARD From Egress Packet Flow Fabric 4 Ingres RX s METRO Queuin 3 2 g Squ CP id U GW From TX Egre Fabric METR ss ASIC O 6 Queu 7 ing 5 Contributing Members
- 18. TIGER: TIME FOR A GREENER INTERNET Switch C Time is Used to Synchronize/Pipeline Forwarding of Time Frames C 3 TF Delay Switch B B Time Frame containing 4 TF Delay a plurality of packets Switch A Contributing Members A TF TF TF TF UTC 0 1 2 3 4 5 6 7 8 9 10 11
- 19. SEASON: SERVICE ENERGY AWARE SUSTAINABLE OPTICAL NETWORKS Multi-fiber, silicon-photonic Bell Labs fast switching & control UNSW End-to-end FEC Contributing Members devices PoliMi Bell Labs Robust & distributed multi-layer control Bell Labs Univ. of Toronto App Center Service-aware flow switching Enterprise Energy & locality aware Energy-aware wavelength placement and execution of routing & protocols app center services Univ. of Toronto INRIA Columbia Univ. CEET
- 20. Thank you! 20 | ALL RIGHTS RESERVED. COPYRIGHT © ALCATEL-LUCENT 2011.