Next Challenges in Optical Networking Research:

odecrackAI and Robotics

Oct 29, 2013 (3 years and 11 months ago)

80 views

Next Challenges in Optical Networking Research:


Contribution from the CaON cluster

Dimitra Simeonidou:
dsimeo@essex.ac.uk
, Sergi Figuerola:
sergi.figuerola@i2cat.net




The
CaON

Vision of Future Optical Networks


Application
driven and technology enabled

High
-
speed data

400G, 1Tb/s

Residential


Media

Flexible Network

MULTI
-
BAND

SSS

Technology
Enabled

Application Driven

Intelligent

Adaptive Optical Networks

MULTI
-
BAND

AMPLIFIER

FAST OPTICAL

SWITCH

SDM

(
DE)MUX

MULTI
-
BAND

SSS

BROADBAND

λ
-
CONVERSION

Elastic use of resources

Flexible use of technology

Cloud


The
CaON

Reference model I


CaON

reference model
presents
a layered architecture
linking optical
networks
with future services and applications


The model promotes the convergence of the optical infrastructure
layers with upper layers and aims to strategically position optical
networks as key enabler of Future Internet and cloud networking
service deployment


The
CaON

Reference model II


A bottom
-
up reference model, where the infrastructure and provisioning
layers, together with cross
-
layer SLA and the management, are
identified the key focus for future research trends within the
CaON

cluster community.


The physical infrastructure layer covers from the core to the access
optical network technologies.

Key Research Challenges for Realizing the
CaON

Reference Model



Support for Multi
-
gigabit Access Rates (FP7 ALPHA, OASIS)


Spectrum management: Flexible, Elastic Optical Layer (FP7
STRONGEST, FP7 call 8 IDEALIST)


Architectures on Demand


Control Plane (FP7 MAINS and STRONGEST)


Targeted extensions for dynamic and data plane
-
aware network services


Software/Hardware Defined Network Programmability (FIRE OFELIA
and FIRE call 8 ALIEN
)


For infrastructure and service adaptation


Optical Network and IT Convergence (FP7 GEYSERS)


Infrastructure
Virtualisation
, Slicing and Isolation


Optical Network Cognition (FP7 CHRON, UK EPSRC Photonics
HyperHighway
)


Energy Efficient Optical Networks (FP7 STRONGEST and TRENT)




Flexible allocation of resources in time and frequency in order to:


Accommodate applications with
arbitrary requirements



Spectrum Management: Elastic Resource Allocation

Video conference/Virtual Presence

Education/Remote Learning

High
-
speed data transmission 400G, 1T

Gaming


Elastic
Time and Frequency plus Space Allocation


Elastic frequency allocation to enable:


Support for high
-
speed channels with arbitrary bandwidth requirements


Better
spectral efficiency for lower bit rates


Elastic time allocation for:


Efficient all
-
optical switching of sub
-
wavelength traffic


Finer
all
-
optical bandwidth
granularities

Continuous channels
at various bit
-
rates

User traffic at various
bit
-
rates and
modulation formats

Space

Novel
Fibres and
Fibre
-
based
components


Optical Networks on Demand


Adapt to traffic profile


Support arbitrary switching
-
granularity


Dynamic Infrastructure Composition (including VI)


Dynamic architecture reconfiguration


Modular infrastructure planning


Seamless integration with other technology domains (network + IT)


Hitless upgrade with new functionality


Wavelength conversion


Regeneration


Optical signal processing


Space division multiplexing (multi
-
core, multimode)


Quantum technologies


Other?




Support of Multi
-
Gbps

Access Rates:



Acceleration of access deployment through


Reduced total cost of ownership


Converged solutions supporting transport of mobile and fixed traffic in both
front
-

and backhaul scenarios


Seamless integration of access and metro/aggregation


Unified control and management planes


Virtualization and context
-
aware networking


New solutions for simultaneous:


More users per feeder (>1000)


Higher speeds (up to 10
Gb
/s peak)


longer reach (100 km)


Green and fast (1
Gb
/s and beyond) home networking



Optical network control plane:



Main research challenges include


True multi
-
vendor and multi
-
carrier control plane solutions, including
extensions for elastic technologies


Split architectures that decouple the control plane from the optical
transport


OpenFlow

as an open/vendor
-
independent interface to network data plane


Multi
-
technology and multi
-
domain path computation services coupled with
traffic optimization


Software Defined Networking at large


Control plane interfaces to external end
-
user “systems” (e.g. clouds) for
any type of bandwidth
-
on
-
demand service and seamless integration
with the service layer workflows.


Optical Network and IT Convergence: for High
Performance, Global Reach Clouds



Provisioning over hybrid infrastructures composed of both IT
resources (i.e. compute, storage, data
centres
) and optical
networks


It will require :


Virtualise

the physical optical network infrastructure (analogue or
digital)


Federate heterogeneous resources from different providers


Unified management and provisioning procedures for the whole
integration with the IT network infrastructures


Specific Issues in Optical Network Virtualization


Optical networks are analogue in nature


More complexity than L2/L3 (digital domain) virtualization as a result of physical layer
impairments and constraints


Slice isolation is a big challenge in optical networks



Physical layer impairments


Affect the isolation between VIs


Newly composed VI
s will affect the existing ones


Affect the ultimate feasibility of VIs



Wavelength continuity constraint


Affect the network resource utilization



Can we use new infrastructure capabilities such as Space Division Multiplexing (multi
-
core?)



Cognitive, self managed optical networks:



Dynamically re
-
purpose, evolve, self
-
adapt and self
-
optimize
functions/devices/systems of the optical network.


Optical/
opto
-
electronic technologies that would allow for environment
-
aware
systems that can change any parameter based on interaction with the
environment with or without user assistance


Cognitive control and management plane for dynamic infrastructure self
-
adaptation across heterogeneous systems.



Energy efficient optical networking:



Improve the design, planning and operations for energy aware
management capable of 100 times energy consumption reduction


Introduction of new simpler protocols


Definition of energy friendly resilience


Support of planning and routing algorithms


Focus on energy efficient optical network services for applications such as
P2P, grid or cloud services