towards newtowards new-generation networksgeneration ... presentaion slides/apnoms2… · towards...
TRANSCRIPT
Towards NewTowards New--Generation NetworksGeneration NetworksTowards NewTowards New--Generation NetworksGeneration Networks-- NICT’s Approach NICT’s Approach --
Masayuki MurataMasayuki MurataMasayuki MurataMasayuki MurataProfessor, Osaka UniversityProfessor, Osaka University
Deputy of NewDeputy of New Generation NetworkGeneration NetworkDeputy of NewDeputy of New--Generation Network Generation Network R&D Strategic Headquarters, NICT R&D Strategic Headquarters, NICT
murata@ist osakamurata@ist osaka--u ac jpu ac [email protected]@ist.osaka--u.ac.jpu.ac.jp
Contents• Vision of NICT towards New-Generation
N t kNetworks– How do we reach the new-generation
k ?networks?– Example of R&D plans for new-generation
networks• Network Architecture for New-Generation
Networks– AKARI at NICT as an architecture exampleAKARI at NICT as an architecture example– Self-organizing networks & bio-inspired
approachesapproaches
M. Murata 2
Why IP Succeeded?
• Hourglass Paradigm– Everything on IP, IP on Everything
email WWW VOIP• KISS Principle– “Keep It Simple, Stupid” by David S.
Isenberg
email WWW VOIP...
SMTP HTTP RTP...
TCP UDP– Today’s optimization is tomorrow’s
bottleneck• Simple network layer: service is
TCP UDP…
IPp yrealized at the end-hosts– Scalable and robust communication– Adaptable to unpredictable new
Ethernet PPP…
CSMA SONET WDMAdaptable to unpredictable new applications
– Source of disruptive innovation
CSMA SONET WDM...
copper fiber radio...
M. Murata 3
Hourglass Paradigm by Deering
IP is Really Simple?
Small Devices
UniversalAccess
UbiquitousAccess
Overlay N t k
email WWW VOIP...
SMTP HTTP RTP
Devices
Necessary functions incrementally added
NetworkSMTP HTTP RTP...
TCP UDP… Mobile IP
HierarchicalAddressing
IP
Ethernet PPPMPLS/GMPLS
IPSecRSVP(-TE)
NAT
Ethernet PPP…
SONET WDM...
copper fiber radio
DHCP
Multicast/Anycast
Cross-LayerOptimization
M. Murata 4
copper fiber radio...AnycastNecessary layers incrementally added
Life Cycle of Innovation
me
ty, I
ncom New Technology
ctio
nalit
Improvement???
ce, F
unc
Current TechnologyReplacement
form
anc
“S-Curve”Everett Rogers, 1962,The Diffusion of Innovations
Complexity, Cost, Time
Per
f The Diffusion of Innovations
5
Incremental Innovation vs. Disruptive Innovation?M. Murata
How to Reach “New Generation Network” NSF FIND/GENIEU FP7
New GenerationNetwork (NWGN)( )
2) Incremental development with a future direction 1) clean-slate design
Virtualization) g
Present Past Next GenerationNetworkNetwork Network (NGN)
6
2005 2010 2015
M. Murata
Constituents for New-Generation Networks
• Establish testbed• Promote testbed-based
i f R&D
• Architecture• Vision and technological
f ibili experiments of new R&D achievements
feasibility• Virtualization• Ubiquitous and sensor
networking gtechnologies
• Wireless• Security
Promote Strategic
R&D Plans
Promote Testbed
Infrastructure
•• Roles of ICT in socio-
Promote International Cooperation
Socio-economical
assessments
• Competition and cooperation
• International
economical problems• Roles for sustainable
society• Future information • International
standardization effortsFuture information society?
7M. Murata
NICT’s New-Generation Network R&D Strategic Headquartersg qhttp://nwgn.nict.go.jp/
New-Generation Network R&D Strategic Headquarters (since Oct. 2007)
Missions• Plans mid-term and long-term new-generation network R&D strategiesg g & g• Internationally promotes R&D strategy and roadmap• Promotes effective and consistent NICT’s R&D• Human-resource development of ICT researchers and engineers
Socioeconomic’s Viewpoints• ICT’s potential against social problems?
Field Technological ItemsEnergy saving/Reduction of Environmental Impact
- Power saving technology- Predict, detect, and alert environmental pollution
Disaster management - Predict, detect, and alert disaster- Protect important data, facilities, and so on
Medical care - HealthcareRemote medical care- Remote medical care
Food shortage - Improve the food-sufficiency ratio- Provide secure food
Secure life/society - Crime preventionSecure life/society Crime prevention- Accident prevention- Secure networks
Gaps - Gaps between cities and countriesSh k l d d i f ti- Share knowledge and information
Aging population - Healthcare- Care for aging people
Globalization Support global mobility
9
Globalization - Support global mobility- Support diversity
M. Murata
Climate Changes
IPCC Fourth Assessment Report (AR4). “Climate Change 2007”, IPCC I t t l P lIPCC - Intergovernmental Panel on Climate ChangeSRC: http://www.ipcc.ch/
10
Crisis of Cultural Diversity
K
French3% Italian
1%
Web Content by Language
h
German6%
Korean1%
Other9%
Japanese6%
Spanish2%
English68%
Chinese4%
11
National Virtual Translation Center (USG) Languages of the Internet http://www.cybertelecom.org/data/content.htm
NICT’s Vision for New-Generation NetworkNew Generation Network
Diversity and Inclusion
Mi i i th N ti
Diversity and Inclusion- Networking the Future -
Natural disasterfood shortage problem
• Minimize the Negatives– To establish new societal systems– Sustainable society – Minimize the environmental impact
gap between countiesglobal temperature change
gap between cities and countriesbl
p– Reliable against the disasters– …
• Maximize the PotentialTo promote increasing wisdom ofenergy problem
aging society with a low birth rateinformation explosion
l ti l i
– To promote increasing wisdom of human-beings
– Improve QOL– Promote innovation
population explosiondigital divide
– …• Inclusion
– Respect diversity of civilization, culture, personnel
– Inclusion for people, inclusion by ICT, inclusion in ICT, planet-scale inclusion, …
12
Energy Problem
• Evidence– Power consumption by ICT account for 5.8% in
2006– Total traffic volume inn 2020 is estimated to be
1000 – 100,000 times larger, g– Social activities utilizing ICT consume more
energyenergy• For sustainable global society to solve the power
consumption problemconsumption problem– New network architecture not consuming the
till b t d li blpower, still robust and reliable13M. Murata
Diversity of Culture, Life, Personnel• Future perspective
Mutual understandings and exchanges– Mutual understandings and exchanges irrespective of differences of culture, society, race
– Allowing diversity, and supporting socialAllowing diversity, and supporting social participation and contribution
• Impact on future society– Getting over cultural barriers – Establish global ubiquitous society
• How ICT can contribute?– Multi-lingual translation technology– Autonomous service technology– Power saving networks
S t ki t h l– Sensor networking technology14M. Murata
Approach for Forming Our Strategy
Diversity and InclusionNetworking the Future
Deep Insights into Technology Value
- Networking the Future -
Minimizing the N ti New GenerationNegatives
Maximizing the
Technology Requirement for New-Generation
New-Generation Network asSocial InfrastructureCreates “New Maximizing the
PotentialNetworks Value”
Inclusion Disruptive Innovation
15M. Murata
Five Challenges for New-Generation NetworkNew-Generation Network
Value-Creation
Autonomicity
Challenge
Energy Challenge
AutonomicityChallenge
Complexity Challenge
HeterogeityChallenge
16M. Murata
Network Architecture Design Project: AKARI ProjectAKARI Project
http://akari-project.nict.go.jp/
EE
Future (possible) requirements from
diverse users and society
Future (possible) requirements from
diverse users and societyFlexible to adopt new i t
Flexible to adopt new i t
Design PrinciplesDesign Principles SCIE
NC
ESC
IEN
CE
No vertical division.common
No vertical division.common Network Architecture
yyrequirementsrequirements
Select, integrate Select, integrate WO
RK
SW
OR
K S
common infrastructure
common infrastructure
Network Architecture
Evolving future fundamentalEvolving future fundamentalCan enjoy fundamentalCan enjoy fundamentalProof -of-conceptProof -of-conceptFeedbackFeedback
and simplifyand simplify
NET
WN
ETW
Evolving, future fundamental technologies
Evolving, future fundamental technologies
Can enjoy fundamental technology advances
Can enjoy fundamental technology advances
TestbedTestbed
conceptconcept
All OpticalAll Optical Software RadioSoftware Radio VirtualizationVirtualization
- Optimal Integration of many components
17
- Stable enough to rely on for a long time
AKARI Sustainable Architecture Principles
CapacitCapacity for
Quality
KISS (Keep It Simple, Stupid)• Crystal synthesis (select, integrate,
simplify)C l (l d )• Common layer (layer degeneracy)
• End-to-end (original Internet)
For F tReliable Future Diverse Society
Reliable Network Space
Reality Connection• ID-Locator separation
Sustainable & Evolutional• Self-* properties (emergent)
M. Murata 18
ID Locator separation• Bi-directional authentication• Traceability
• Autonomic distributed control• Scalable• Social Selection
AKARI Architecture FocusAKARI Photonic Switching - Path / Packet Integration -
Q S G t d Requirements:QoS Guaranteed Requirements:- Handle different types of traffic- Peta-bps switching capacity- Tera-bps link speed (40G x 100)- 100 billion tiny terminals
ectr
um
Header Payload
Optical Circuit Switch Parallel Optical Packet Transmission
y
A,B,C,D E,F,G,H,I,J,K,L,M,N,O,P,Q,R,S,T
Wavelength
Pow
er S
pe Header Payload
Optical Wave , , ,
Sustainable in Capacity
Optical WaveSharing Optical Packet Switch
(with Fiber Delay Line)- Scalable
E S i
Keep Both QoS Assuredness & Best Effort Efficiency=> OPS + OCS (Not OPS over OCS)
- Energy-Saving
(C) National Institute of I f ti d C i ti
19
=> OPS + OCS (Not OPS over OCS)Best Effort Sustainable in Usage - Allow Usage Diversity
AKARI Architecture Focus:Self-Organizing NetworksSelf-Organizing Networks
• Self-organization is a set of dynamical mechanismsdynamical mechanisms whereby structures appear at the global level of a system from interactions among its lower-
Simultaneous failures
Simultaneous failures Bad Bad interactions among its lower-
level components. The rules specifying the interactions among the system's constituent
failuresfailures
ChainChainfailures Software Software
provisioningprovisioning
among the system s constituent units are executed on the basis of purely logical information, without reference to the global
failuresfailures bug
without reference to the global pattern which is an emergentproperty of the system rather than a property imposed upon OptimalityRobust
Efficiency
p p y p pthe system by an external ordering influence. Adaptability
E. Bonabeau, M. Dorigo, G. Theraulaz, Swarm Intelligence: From Natural to Artificial Systems, Oxford University Press, 1999.
20M. Murata
Four Principles for Self-Organization
• Four principle mechanisms for self-organization in bi l i l t
Positive feedback:permits evolution and
biological systems:
permits evolution and promotes creation of
structure (reinforcement)
Direct or indirect interaction among
Utilization of inherent randomness and interaction among
individualsrandomness and
fluctuations
Negative feedback: regulates influences from
i b d d t tiprevious bad adaptations (saturation, competition)
M. Murata 21
Technology Requirements for New-Generation Network ArchitectureNew Generation Network Architecture
Continuously growable
Adaptive and self-organizing network Topologically-
changing network(sustainable) network changing network
Scalable network Real-time traffic Scalable network control measurements
Dynamic Interactions between and within layers
Interested in the behavior of whole system, including vertical and horizontal relations
M. Murata 22
including vertical and horizontal relations
Metrics for New-Generation Networks?New-Generation Networks?
• “Beyond capacity”• Beyond capacity– “*-ties” other than mean delay, throughput, packet loss probability…
ReliabilityComplexity Availability
Manageability
Adaptability
Reconfigurability
Dependability
ReconfigurabilitySustainability
• “Quantity-rich network” to “quality-rich network”
M. Murata 23
New Approach for Designing New-Generation NetworksGeneration Networks
Traditional approach
New approach
Traditional approach
i bilit
Technologically improvement
ithi f
rman
ce survivability↓
sustainability
within a few years
perf
or ↓ dependability
# of simultaneous failures, degree of environmental changes/influences of failures、
Self-organizing networks based on bio-inspired approachesOur approach
M. Murata 24
g g p ppPrinciples: interaction, feedback, randomness
Bio-inspired Network Control• New emerging networks have unique characteristics
different from existing wired networksdifferent from existing wired networks• Expect to learn robustness, adaptability and self-organizing
properties of biological systemsBi l i t i t t th d t bilit t th h i i t– Biologists point out the adaptability to the changing environments, and as a result robustness in the biological systems is excellent.
• while, it is rather slowIncorporate the self organized and autonomous mechanism in– Incorporate the self-organized and autonomous mechanism in biology into the communication network
Ref The behavior of natural systems may appear unpredictable andRef. The behavior of natural systems may appear unpredictable and imprecise, but at the same time living organisms and the ecosystems in which they live show a substantial degree of resilience. (“Toward Self-Organizing, Self-Repairing and Resilient L S l Di t ib t d S t ” A M t t l T h i lLarge-Scale Distributed Systems,” A. Montresor et al. Technical Report UBLCS-2002-10, Sept. 2002.
M. Murata 25
Swarm Intelligence• The emergent collective intelligence of
groups of simple agentsgroups of simple agents • Interaction in Biological Groups
– A social group exhibits an intelligent g p gand organized behavior without any centralized control, but with local and mutual interactions amongand mutual interactions among individuals
– Interactions can be: direct or i di ( i )indirect (stigmergy)
– The behavior is adaptive to changes in the environmentin the environment
– The group keeps working even if an individual fails
26M. Murata
Bio-inspired ExamplesOverlay Network Symbiosis
symbiosis of different cells, organisms, groups, and species
Waveform Synchronized Data Gatheringsynchronized flashes in a group of fireflies
Reaction-Diffusion based Control Scheme for Sensor Networks
synchronized flashes in a group of fireflies
Scheme for Sensor Networkspattern formation on the surface of the body of
an emperor angelfish
Scalable Ant-based Routing Schemeforaging behavior of ants
Scalable Congestion Control for Transport Layer Protocol
Lotka-Voltera model based on ecosystems
27M. Murata
Lotka Voltera model based on ecosystems
For detailed information, visit at http://www.anarg.jp/
Summary• Vision of NICT towards New-Generation
N t kNetworks– How do we reach the new-generation
k ?networks?– Example of R&D plans for new-generation
networks• Network Architecture for New-Generation
Networks– AKARI at NICT as an architecture exampleAKARI at NICT as an architecture example– Self-organizing networks & bio-inspired
approachesapproaches
M. Murata 28