Top Banner
Towards Dynamic and Scalable Optical Networks Brian Smith 3 rd May 2005
28

Towards Dynamic and Scalable Optical Networks Brian Smith 3 rd May 2005.

Dec 15, 2015

Download

Documents

Sophia Frake
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Towards Dynamic and Scalable Optical NetworksBrian Smith 3rd May 2005

Page 2: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Towards Dynamic and Scalable Optical Networks

What is required to deliver truly dynamic optical networks ?

A Dynamic Control Plane

Technologies for Wavelength Switching

Considerations for achieving higher data rates + capacities. Issues with 40Gbps

Is 100Gbps achievable ?

Getting maximum spectral efficiency in a dynamic optical network.

Page 3: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

A Dynamic Control Plane

Page 4: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Lightpath Control

To deliver on-demand gigabit lightpaths, a fast and reliable distributed control plane is required Reliable –transport infrastructure should remain stable

during reconfigurations

Fast – otherwise its not on-demand !

GMPLS is one control plane under development for optical networks by the industry. Based on standard set of IP routing and signaling protocols

UCLP is an example of an R&E initiative (CANARIE) TL1 interfaces controlled by a distributed service layer

based on a web browser and network model.

Page 5: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Evolving Towards Dynamic Lightpath Control

Features Required for Light-path Control

Topology discovery and link management.

Operator signaled light-paths (Network should automatically manage its own demand).

Client on-demand light-paths (High end users can individually control wavelengths). An important feature for future R&E networks !

Integration with IP/MPLS control plane for dynamic traffic engineering e.g HOPI / DRAGON.

Dynamic protection – if required

Page 6: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Backbone

Network

On-Demand Bandwidth Capacity

AccessNetwork

MeshNetwork

High EndUsers

A B

C

D

Server Farm

User ControlledLightpath(e.g. for nightly databack-up)

Available now between routers. Needs to evolve to support high data rates on wavelengths

Wavelength Switch Routers

Page 7: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Example of need for on-demand wavelengths

4 radio telescopes in an array – 12 hour observation

Assuming 1Gbps per telescope – 0.2 Petabits of data !

How long would it take to back up the data to storage ?

With 100Mbps rate – ~23 days minimum assuming no packet loss.

With dedicated GigE wavelength – ~4 days.

If user can request an on demand 10GigE wavelength – ~5 hours.

High-end Research users will require high capacity on demand services

Page 8: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Enabling Technologies For Dynamic Networks

Page 9: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Enabling Technologies for Dynamic Networks

Electronic ROADM

Optical ROADM

Tunable lasers Tunable 10G DWDM XFPs will be available in 2006

Integrated optical wavelength converter / tunable laser Demonstrated in Labs using non-linear cross-talk in Semiconductor

Optical Amplifiers – Capable of supporting up to 40Gbps

Page 10: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Electronic ROADM

NxN

TransparentWavelength

Switch(electrical)

Trib1310

Trib1310

Trib850

Trib1550

DWDM

WestFiber

EastFiber

DWDM

DWDM

DWDM

NorthFiber

CWDM

CWDM

SouthFiber

CWDM

CWDM

• Native signal transparency with layer 1 performance monitoring

• Simple Any-to-Any Multi-Degree grid interconnection

• Simple to Engineer.

Page 11: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Optical ROADM – Wave-blocker

Splitter Wave-blocker

Drop Filter

Add Filter

Coupler

• Drop and Add Filters must be tuneable for maximum flexibility.

• Hitless filter tuning is a problem.

• Many discrete components so expensive

• High insertion loss – Limits DCM – Limits reach between nodes for fully transparent networks.

Page 12: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Optical ROADM – Wavelength Selective Switch (WSS)

WavelengthSelective Switch

Add

Coupler

DropChannels

OptionalExpansionPort

• Fewer discrete optical components

• Fully flexible colourless add/drop

• Lower insertion loss

• Limited number of drop ports – Use expansion port !

Page 13: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Comparison- Wavelength Switching

Functionality Electronic ROADM Optical ROADM

Transparency(bit rate and protocol)

Yes- wide range of signals

Yes

Low Latency Yes Yes

Single wavelength granularity(I.e. no wavelength stranding)

Yes Yes

Mesh Support (multi-degree) Yes Yes- Blocking issues

Wavelength Translation Yes No

Grid Conversion(e.g. CWDM to DWDM)

Yes No

Protocol Performance Monitoring

Easy Optical power only

Wavelength Protection & Hitless Maintenance

Easy Ring – Easy

Mesh – More difficult

Page 14: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Implementing ROADM Interfaces

Optical and Electronic ROADM complement each other.

Trib1310

DWDMNxN

TransparentWavelength

Switch(electrical)

WestFiber

EastFiber

CWDMWest

Trib1310

Trib850

Trib1550

CWDMEast

DWDM

CWDM

CWDM

DWDM

DWDM

CWDM

CWDM

Pass-through TrafficNorthFiber

DWDM

DWDM

DWDM

DWDM

SouthFiber

OpticalROADM

I/F

OpticalROADM

I/F

Page 15: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Multi-Degree ROADM Interfaces

First step towards full NxN photonic wavelength switch.

Optical Pass-through channelsNorthFiber

SouthFiberOptical

ROADMI/F

OpticalROADM

I/F

NorthWFiber

DWDM

NxN

TransparentWavelength

Switch(electrical)West

FiberEastFiberDWDM

DWDM

DWDM

OpticalROADM

I/F

SouthEFiberOptical

ROADMI/F

DWDMDWDM DWDM

DWDM

Page 16: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

2.5Gbps - 2 DEGREE NODE

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1 6 11 16 21 26 31 36

# PASSTHRU WAVELENGTHS

CO

ST

(A

U)

ELECTRICAL OPTICAL

~17

Cost Comparison – 2.5G Traffic

Page 17: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

10Gbps - 2 DEGREE NODE

0

0.5

1

1.5

2

2.5

1 6 11 16 21 26 31 36

# PASSTHRU WAVELENGTHS

CO

ST

(A

U)

ELECTRICAL OPTICAL

~6

Cost Comparison – 10G Traffic

Page 18: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Wavelength Switching - Cost sweet spots

4 8 12 16 20 24 28 32Pass-through Channels

Optical ROADM

ElectronicROADM

Optical ROADM

ElectronicROADM

10G

2.5G

ChannelRate

Note:For 2-degree metro ring applications. Also applies to 4-degree mesh architecture

Page 19: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

The Future of 40G/100G

Page 20: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

40Gbps/100Gbps

40Gbps 40Gbps DWDM trials and demonstrations becoming more common.

Ability to overlay on existing 2.5/10G links – a key driver !

40Gbps router interfaces have been demonstrated.

Dispersion must be controlled within ± 62 ps/nm.

PMD is an issue. Cannot exceed 2ps (outage < 3min/year)

100Gbps Can 100Gbps be achieved over DWDM ?

Dispersion tolerance even tighter - ± 25 ps/nm.

PMD more of an issue. Cannot exceed 1ps (outage < 3min/year)

Page 21: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

40Gbps Dispersion Tolerance 6x80kmx26dB - 32

• 100GHz spacing SPM, XPM and FWM effects included

Range of possible net dispersion

0

1

2

3

4

5

6

-600 -400 -200 0 200 400 600

Net Dispersion (ps/nm)

Pen

alty

(d

B)

40Gbps 10Gbps

Tunable Dispersion Compensation Required for 40Gbps.

Page 22: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

100Gbps

Several published examples of single wavelength 100Gbps+ transmission.

Spectral width ~ 150 GHz for NRZ so won’t fit into a 100GHz spaced DWDM pass-band (~85GHz) !

Dispersion limit for NRZ is ± 25ps/nm.

If we use non-binary coding – Spectral width reduced to 75GHz – Just fits within 100Ghz spaced DWDM band.

Needs tight control of laser + filter wavelengths.

Using >1 bit per symbol coding technique such as duo-binary or QPSK improves tolerance to dispersion and PMD.

100Gbps is achievable. Needs sophisticated coding!

Page 23: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Polarization Mode Dispersion

Using 6x80kmx26dB with 6 EDFA and 6 DCM, the calculated average DGD (assuming fiber is post 1995) = 2.5 ps

The PMD tolerance (and expected outage) for various data rates is:

Rate pmd tolerance system outages/yr

2.5G 30ps insignificant pmd outages/yr

10G 7.6ps insignificant pmd outages/yr

40G(NRZ) 2ps ~ 3 minutes/year assuming FEC

100G(NRZ)0.9ps Requires PMD compensation

Page 24: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

How Much Capacity ?

100Gbps

Duo-binary

Wave-locker++

1b/s/Hz

16 symbol levels – 4 bits per symbol required.

256 symbol levels – 8 bits per symbol required.

40Gbps

NRZ/CS-RZ/

Wave-locker+

10G overlay

0.4b/s/Hz

Duobinary

Wave-locker+

0.8b/s/Hz

16 symbol levels – 4 bits per symbol

10Gbps

No issue

NRZ

0.1b/s/Hz

Reduced reach

Wave-locker

NRZ

0.2b/s/Hz

Reduced reach

No ROADMs

Wave-locker+

0.4b/s/Hz

100GHz 50GHz 25GHz

Page 25: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Summary

On-Demand Light-path Control Enabled by:

Distributed, Intelligent Light-path Control (UCLP, GMPLS)

Electronic and Optical ROADM.

Widely Tunable Laser Sources.

40Gbps/100Gbps 40Gbps can be deployed over existing 10G infrastructure with

appropriate dispersion control + FEC.

100Gbps will a challenge requiring sophisticated coding schemes and components for PMD mitigation.

Page 26: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Thank You

Page 27: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Impact of Tighter Channel Spacing

Four Wave Mixing (FWM)

-30.00

-25.00

-20.00

-15.00

-10.00

-5.00

0.00

0 20 40 60 80 100 120

WAVELENGTH SPACING (GHz)

FW

M E

FF

ICIE

NC

Y (

dB

)

SMF-28 LEAF DSF

Increased FWM Impact – Reduced Reach.

Page 28: Towards Dynamic and Scalable Optical Networks  Brian Smith 3 rd May 2005.

Impact of Tighter Channel Spacing

Cross Phase Modulation distortion (XPM)

-20

-18

-16

-14

-12

-10

-8

-6

-4

-2

0

0 20 40 60 80 100 120

WAVELENGTH SPACING (GHz)

XP

M E

FF

ICE

NC

Y (

dB

)

SMF-28 LEAF DSF

Increased XPM Impact – Reduced Reach.