Networking in DWDM systems. Péter Barta András Kalmár 7-9. of April, Debrecen

Transcription

1 Networking in DWDM systems Zero Touch Photonics Péter Barta András Kalmár 7-9. of April, Debrecen

2 Outline Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 2 Zero touch photonics April 2010

3 Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 3 Zero touch photonics April 2010

4 Transformation Drivers for Photonic Networks New services (and new players) are emerging, requesting huge bandwidth growth in photonic networks Fixed and mobile, personalized Increased popularity of new Video and Internet applications (> 6 x growth between 2004 and 2008) Service providers are facing new challenges React faster, providing better time-to-service Higher flexibility and capacity Control costs Highly automated operations and resilience Triple Play bundle IPTV VoIP (incl. HDTV, VoD)! More Bandwidth Blended, personalized voice, video & data services Enhanced Internet Services today Coming soon 4 Zero touch photonics April 2010

5 Challenges Increased traffic demands mean: Significant growth of the number of the services to deploy within the WDM network (more and more difficult to handle, to administrate, etc.) More dynamic nature of the services More diverse end points of the services To meet these challenges networking approach shall be introduced in the DWDM networks instead of the static systems deployed today. 5 Zero touch photonics April 2010

6 Static Networks Based on fixed Wavelength filters Typical WDM Network today OADM Ch 1-8 Ch 1-8 HUB Ch 9-16 Ch Ch Physical WDM Ring Ch OADM Ch 9-16 OADM Ch OADM 6 Zero touch photonics April 2010

7 Static Networks Based on fixed Wavelength filters Topology and capacity/node determined at time of network design Traffic projection based upon best estimates at the time Frequent changes even during design/bid/deployment Not always cost effective to modify the system Can lead to premature system exhaust Expected system life time: 5 10 years No accurate traffic projections available for such a long period Insufficient No. of wavelengths available to hot spots Unlit wavelengths to cold spots cannot be utilized Topology is inconsistent for emerging applications Telephony, SAN, Enterprise, VoIP topologies looks different 7 Zero touch photonics April 2010

8 Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 8 Zero touch photonics April 2010

9 Transforming WDM Networks Zero-Touch Transparent Photonic Networks Flexibility Eliminates need to forecast traffic Multidegree Meshed ROADM, tunable lasers and filters, Performance Eliminates regeneration points Multi-degree, Gain Equalization, New modulation formats, 40G/100G, Raman, Automation Eliminates manual intervention Auto power adjust., GMPLS/ASON restoration, OTH, smart photonic manager, Integration Eliminates multiple equipment C+DWDM, multireach, GE ADM, L2 switching, True Networking Wavelength management and switching OAM and Survivability Reduced TCO Improved time-to-service Zero-Touch Transparent Photonic Networks transform WDM into true transport networking for simplified and accelerated operations 9 Zero touch photonics April 2010

10 Technology Enablers for Zero-Touch Photonic Networking Beyond Multi-Degree ROADM WSS T&ROADM Unique Architecture Traditional ROADM Architecture Wavelength Tracker TOADM Full Tunability Color-less ports E2E Traffic Management Path, QoS, power monitoring Photonic Restoration GMPLS/ASON Re-coloring ROADM Wavelength routing at intermediate nodes Still need to go to sites for the add/drop Multi-degree Increased connectivity (mesh) Easier design Still need to go to sites for the add/drop Multicasting Wavelength routing e2e Full Flexibility Simple design, planning (highly reduced blocking), operations and inventory No need to go to sites Misconnections management Full Operations Touch-less design, planning and operations NOC based automated commissioning, supervision and operations Full Automation Low cost-per-bit photonic bandwidth resilience, no blocking, more resources Complement digital survivability (TDM/OTN, packets) Touch-less, flexible networking layer for TCO control WSS: Wavelength Selective Switch; T&ROADM: Tunable ROADM 10 Zero touch photonics April 2010

11 Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 11 Zero touch photonics April 2010

12 Flexibility: a key enabler Ease network operation west Tunable ROADM east Color-independent add drop: Any color to any port More transparent directions: Ł less OEO and sub-λ switching Route-fixed add drop Degree-N Route-fixed add drop Gbit/km cost reduction west ROADM east Degree-N colored add drop Automated network provisioning and system tuning add drop reroute Ł restoration Improve network resilience 12 Zero touch photonics April 2010

13 TOADM architecture AMP IN Filter (East) THRU Filter (West) AMP OUT (Optional) WDM IN WSS WSS WDM IN 8 Colorless ports 8 Colorless ports AMP IN AMP OUT (Optional) Transponder Transponder Transponder Transponder Scalable TOADM/ROADM/N-degree node architecture 13 Zero touch photonics April 2010

14 N-Degree DWDM node architecture AMP IN Filter (East) THRU Filter (West) AMP OUT (Optional) WDM IN WSS WSS WDM IN BB port 6 colorless ports BB port 6 colorless ports AMP IN AMP OUT (Optional) SFD44 Xpdr SFD44 Xpdr Xpdr Xpdr Xpdr Xpdr AMP OUT (Optional) SFD44 Xpdr SFD44 Xpdr WDM IN BB port 6 colorless ports BB 6 colorless ports port WSS WSS AMP IN WDM IN AMP IN Filter (North) THRU Filter (South) AMP OUT (Optional) 14 Zero touch photonics April 2010

15 T&R-OADM: value proposition Benefits of the Tuneable add-drop functionality: 1. Set-up or re-engineer a circuit within minutes Non-tuneable ROADMs avoids intervention at intermediate sites, but still require intervention at circuits endpoints Tuneable architectures remove intervention at circuits endpoints (provided that transponders are available) 15 Zero touch photonics April 2010

16 The need: re-engineering a circuit Initial link set-up: existing circuits A-D and B-C TR-OADM TR-OADM TR-OADM TR-OADM A B C D New need: change circuits to A-C, B-D 16 Zero touch photonics April 2010

17 Standard approach: non-tuneable ROADM R-OADM R-OADM R-OADM R-OADM A B C D Non-Tuneable ROADM architecture: Go to site D, and change transponder cabling Go to site C, and change transponder cabling Activate new service 17 Zero touch photonics April 2010

18 Tuneable ROADM: re-engineering a circuit within minutes TR-OADM TR-OADM TR-OADM TR-OADM B C A D Tuneable ROADM architecture: Just re-program your DEMUX ports, no site intervention needed! 18 Zero touch photonics April 2010

19 T&R-OADM: value proposition Benefits 1. Set-up or re-engineer a circuit within minutes Non tuneable architectures avoid intervention at intermediate sites, but require still intervention at circuits endpoints Tuneable architectures allow to avoid intervention at circuits endpoints (provided that transponders are available) 2. Make channel planning unnecessary All items are colorless, i.e. not associated to a specific wavelength 19 Zero touch photonics April 2010

20 T&R-OADM value proposition: Make channel planning unnecessary Initial traffic plan: Dual homed traffic matrix with minimum frequency usage A B C D E F New need: Install a new service A-B and B-F, A-C and C-F Zero touch photonics April 2010

21 T&R-OADM value proposition: Make channel planning unnecessary Non-tuneable ROADM: Deploy new set of MUX/DEMUX, as the green MUX-DEMUX channels are fully used Non Tunable R-OADM Non Tunable R-OADM Non Tunable R-OADM Non Tunable R-OADM Non Tunable R-OADM Non Tunable R-OADM A B C D E F Fixed-color add/drop ports NON-tuneable MUX/DEMUX 21 Zero touch photonics April 2010

22 T&R-OADM value proposition: Make channel planning unnecessary Tuneable ROADM: Just re-program the tuneable MUX ports to any color in the C+ band TR-OADM TR-OADM TR-OADM TR-OADM TR-OADM TR-OADM A B C D E F Programmed ports Tuneable MUX/DEMUX Unprogrammed ports 22 Zero touch photonics April 2010

23 T&R-OADM value proposition: Make channel planning unnecessary Simulation: Graph: Ring network Any-to-any traffic: Random traffic evolution, from 1 to max capacity, New channel assignment decided at each step without any knowledge of the future demands X-axis: # of nodes in the ring Y-axis: MUX port utilization in %, at full-channel loading MUX port utilization (%) T&R-OADM Non-tuneable R-OADM T&R-OADM up to 3.5 times more efficient than non-tuneable R-OADM! Number of nodes in the ring 23 Zero touch photonics April 2010

24 T&R-OADM: value proposition Benefits 1. Set-up or re-engineer a circuit within minutes Non tuneable architectures avoid intervention at intermediate sites, but require still intervention at circuits endpoints Tuneable architectures allow to avoid intervention at circuits endpoints (provided that transponders are available) 2. Make channel planning unnecessary All items are colorless, i.e. not associated to a specific wavelength 3. Dramatically simplify logistics Reduce the lead-time to define an order (no longer any need to check the installed base) Eliminate possibilities of errors when ordering new items (wrong transponders and MUX colors is a typical cause of errors) Simplify and optimize spares management 24 Zero touch photonics April 2010

25 CAPEX benefits: Degree-N transparency AVERAGE NODE CONNECTIVITY = 3.1 Source: Alcatel-Lucent study Network studies show ~30% CAPEX savings with Transparent OOO nodes European national network Map of physical links and traffic demands With O-SNCP protected traffic APE 14 spans -28% Transparent degree-n OOO nodes APA 20 spans Degree-2 ROADMs WB/WSS Amplifier MUX TPD Remove external nodes (e.g. SDH, Routers,..) for Ring Interconnections 25 Zero touch photonics April 2010

26 The next step: Wavelength rerouting to increase the overall network reliability with T&ROADM: Local OTS Port: Towards photonic transparent networks The local OTS ports represents the ultimate flexibility: local add/drop traffic is not dependent on any specific direction Paths can be provisioned in the photonic domain The enabler for photonic restoration! 1 L 2 Direction 2 pool of regenerator on local OTS port can be accessed dynamically for new connections 3 26 Zero touch photonics April 2010

27 Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 27 Zero touch photonics April 2010

28 Colored, direction dependent nodes (ROADM network) / 1 Initial path configuration 28 Zero touch photonics April 2010

29 Colored, direction dependent nodes (ROADM network) / 2 Remote cfg on intermediate nodes NMS Rerouting for maintenance activity 29 Zero touch photonics April 2010

30 Colorless, direction dependent nodes (TOADM) / 1 l 1 l 1 Initial network condition 30 Zero touch photonics April 2010

31 Colorless, direction dependent nodes (TOADM) / 2 l 1 l 2 Solving possible wavelength contention issues l 2 l 1 Rerouting for maintenance activity with wavelength reallocation 31 Zero touch photonics April 2010

32 Colorless, direction dependent nodes (TOADM) / 3 l 1 l 2 Solving possible wavelength contention issues l 2 Remote wavelength reconfiguration NMS l 1 Remote cfg on intermediate nodes NMS Rerouting for maintenance activity with wavelength reallocation 32 Zero touch photonics April 2010

33 Colorless, directionless nodes (TOADM) / 1 l 1 Rerouting for maintenance activity l 1 Initial network condition 33 Zero touch photonics April 2010

34 Colorless, directionless nodes (TOADM) / 2 l 1 l 1 l 1 Remote cfg on intermediate nodes NMS Rerouting for maintenance activity 34 Zero touch photonics April 2010

35 Automated photonic layer: GMPLS for fast service setup (BW-on- Demand), restoration Switched Connection restoration supporting multiple failures Source-based restoration mode : photonic routing engine for optical route choice GMPLS benefits: Faster service setup Automatic restoration High level of resilience Good resources usage Automated discovery UNI Automatic check Optical Path (direct or regen) UNI OCh NE OCh NE GMPLS RSVP-TE OCh NE OCh NE OOO Transparent Domain UNI OCh NE UNI RSVP-TE OCh NE IP/MPLS router Multi-Service XC degree-n node 35 Zero touch photonics April 2010

36 Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 36 Zero touch photonics April 2010

37 The boundary condition: Spectral efficiency is key Network capacity is growing Higher spectral efficiency? Wider amplification bands? Light another fiber? What is lowest cost? Support legacy systems 50-GHz channel spacing Multiple ROADMs Trade regeneration cost with transparent reach Cost / complexity of transponders Optically-routed network impairments: ROADMs, PMD, CD, Nonlinearities [Magill, LEOS 07] 37 Zero touch photonics April 2010

38 Scalability: 100 Gigabit Ethernet transport Moving to 100GE is to increase capacity, so ŁLink capacity needs to be increased accordingly* ŁIn core networks, scalability of fiber links clearly advocate for higher spectrum efficiency and single wavelength solution with sophisticated modulation format. Historically at 2.5G, typ. 32ch: b/s/hz (100 GHz) Typ. now at 10G: 0.2 b/s/hz (50 GHz grid) Typ. now at 40G: 0.8 b/s/hz (50 GHz grid) *Reminder: today s typical link capacity is ~1 Tbit/s (100x10G) Target for 100G: better spectral efficiency compared to 40G ~2 b/s/hz is achievable (50 GHz grid) Status: Substantial market interest for 100G Alcatel-Lucent Technology leadership including standards development 38 Zero touch photonics April 2010

39 Drivers Flexibility and transparency benefits Tunable ROADM (Reconfigurable Optical Add/Drop Multiplexer) N-Degree T-ROADMs (N-Directions) Optical resilience and restoration System automation System scalability Management & planning 39 Zero touch photonics April 2010

40 Wavelength Tracker Encode/Decode Points WDM IN AMP IN Filter (East) THRU Filter (West) AMP OUT (Optional) AMP OUT (Optional) A A AMP IN WDM IN ADD/DROP = Wavelength Tracker Insertion Point = Wavelength Tracker Detection Point 40 Zero touch photonics April 2010 Transponder v Each channel is encoded with a unique WaveKey pair that allows the channel to be identified and its power monitored. v WaveKeys are encoded onto the channel at each Transponder and act as Optical J0 Trace Identifiers. v The assignment of WaveKeys is managed by the NEs, which maintain a database of the WaveKeys used in the network. v Wavelength Tracker WaveKeys are detected on the following cards: Amplifiers Wavelength Routers v At each detection point, the WaveKeys are detected and their power measured. Transponder

41 Wavelength Tracker Optical Power Management Features: v Wavelength path trace v Fault sectionalization & isolation v Remote optical power control v Threshold alarming v Automated fault correlation Optical Fiber View Management of all wavelength on a selected fiber Optical Path View Management of all points along one wavelength Service-aware management of the optical layer 41 Zero touch photonics April 2010

42 Integrated Alcatel-Lucent Solution: Greater intelligence, Accelerated deployments Planning Tool Equipment configuration automatically generated User-friendly interface with graphical views of the network Plan Current configuration is imported to LPT to support upgrade of existing design Manage Deploy Validated network design exported to NMS Network Management Graphical toplogical view End to and Service awarness 42 Zero touch photonics April 2010

43 Conclusion Zero-Touch Transparent Photonic Networks help service providers to: Ease operations Tunable ROADM, to ease commissioning Flexible nodes, to provision/reconfigure and tune remotely Reduce costs Cost-effective scalable Multi-degree ROADMs enable transparent transits Improve network resilience Photonic restoration Supporting multiple failures in addition to fast electrical/client protection Cap existing, future-proof investments Photonic switching to support variable grid and channel bandwidth Future-compatible (40G/100G ready) and backward-compatible Ensured operation Wavelength Tracker TM and graphical end to end management allow fast and accurate reactions 43 Zero touch photonics April 2010

44 44 Zero touch photonics April 2010