LW_20070501_May_2007

23 Centre Street New Bedford, MA USA 02740-6322Toll Free: 800-IS-FIBER / Tel: 508-992-6464 / Fax: 508-991-8876

e-mail : [email protected] Please visit us on the web: WWW.FOCENTER.COM

Here at ngstrmBond, goal #1 is to make life easy on our customers.No other adhesives manufacturer can offer the level of application

and technical support that we can. None. ngstrmBond was developedby people within the fiber optics industry, for people in the fiber optics

industry, so we understand your bonding needs. We offer:

ngstrmBond thinks "outside the box" to provide solutions for yourindividual needs and applications. Custompackaging, such as 2 - 100 gram

packets, syringes and cartridges, custom formulations, custom colorations,

different viscosities and other properties are what we specialize in. We've

done the homework, so you don't have to. So whether you are in the

Fiber Optics, Optics, Aerospace, Deep Space, Electronics, Medical, or

Rapid Prototyping Industries, ngstrmBond makes it easy to obtainthe best adhesive solutions, so you get the results you're looking for.

Contact us to see how easy it is to get theWorld-Class support you deserve.

ngstrmBond, the only adhesive line developed exclusively for fiber optics.

ngstrmBond Offers the bestCustomer Servicefor Specialty Adhesives

Only

Getting World-Class Support is as easy as pressing this button

Shipment from Stock

Production Volume Capacity

Industry-proven products, such as AB9112, AABB9113SC, AB9123 & ET353ND

Previous Page Contents Zoom In Zoom Out Front Cover Search Issue Next Page BA

M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

____________


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

________________________________

Focus on

page 13

OPTICAL TECHNOLOGIES, COMMUNICATIONS APPLICATIONS, AND INDUSTRY ANALYSIS WORLDWIDE MAY 2007

page 23

www.lightwaveonline.com $23.00

APPLICATIONS

Service-enabled networks aid convergenceBy ANDY McCORMICK

Today, network operators con-tinue to seek convergence of their networks to reduce op-erational expenses and elim-inate unnecessary additional capital expenditures on multi-ple parallel networks. Driven by the need for rapid returns on investments (ROIs), tele-com carriers are integrating a variety of services includ-ing residential broadband, advanced enterprise data con-nectivity, and video into ex-isting voice networks while cable operators are adding voice over IP (VoIP) and ad-vanced enterprise data ser-vices to the existing video and residential broadband

infrastructure. At the same time, there is a massive migra-tion from traditional SONET/SDH-based TDM services to IP/Ethernet-based services, including voice, video, Inter-net access, and Layer 2 busi-ness services.

The migration is not with-out its challenges and risks, especially when you consider supporting converged ser-vices without stranding in-vestments. This task requires the development of a service-enabled transport network de-signed to handle VoIP, video, and data traffic as well as stan-dard voice without sacrificing the performance network op-erators have

INDUSTRY

EDC vendors try again on LRMBy STEPHEN HARDY

Critics of the 10GBase-LRM often warned of the difficulty of relying on electronic dis-persion compensation (EDC) to push a serial 10-Gbit/sec data stream down 220 m of leg-acy 62.5-m multi-mode fiber. Judging by how long it has taken X2 transceivers compliant with the LRM specifications to reach the marketand the fact that most of the first generation of these modules use an EDC chip from the same vendor, de-spite several supplier optionssuch pre-dictions apparently proved sound. Nev-ertheless, the chip vendors have not given up. While licking the wounds generated during the first X2 skirmish, IC suppliers have begun to retool their offerings in hopes of earning fu-

ture design wins, either for a cost-reduction design round in the X2 space or for the first generation of SFP+ applications.

The 10GBase-LRM PHY was designed to of-fer a serialand, hopefully, cheaperalterna-

tive to four-channel 10GBase-LX4 mod-ules for 10-Gigabit Ethernet applications over legacy fiber in en-terprise and data cen-ter applications. LRM proponents touted EDC as the key to overcoming the dis-persion inherent in such multimode fiber that previously had prevented practical serial transmission. However, difficul-

ties soon became apparent when the working group charged with creating the specifications decided to abandon matching the 300-m reach of the LX4 in favor of a more

TECHNOLOGY

T-MPLS and PBT/PBB-TE offer connection-oriented packet transportBy DANIEL JOSEPH BARRY

Provider Backbone Transport (PBT) and transport MPLS (T-MPLS) are two new packet-transport concepts that have re-ignited discussion on the best approach to implement a fully packet-based telecom-munication network. Both concepts propose a connec-tion-oriented, point-to-point packet transport method that provides a high-level of pro-tection, monitoring, and con-trol for network operators.

T-MPLS has been proposed and specified by several vendorsin-cluding Alcatel, Ericsson, Fujitsu, Huawei, Lucent, and Tellabsand standardized by ITU-T over the past two years. PBT, on the other hand, has been proposed by Nortel and is expected to begin standard-ization in 2007 within IEEE as PBB-TE (Provider Back-bone Bridging-Traffic Engi-neering). Other vendors such as Extreme, Meriton, Siemens,

and World Wide Pack-ets have recently voiced

their support.The main appeal of these

concepts is the promise of lower cost, lower complexity, and ease of management, while providing an approach suitable for IP-based services that re-sembles existing, well-under-stood transport technologies.

Migration from SONET/SDHThe develop-

Scale model

How do we get to cost-effective 100GbE? PAGE 13

Going global

Carriers seek mul-tiservice Ethernet platforms. PAGE 23

Pulse check

The industrys recovery is underway. PAGE 31

Deep impactPatterning a filter via deep ultravi-olet lithography helps LightSmyth create a nanoscale photonic component with flexible passband control. PAGE 20

page 31

While Vitesse cleaned up in the initial round of 10GBase-LRM X2 designs, a subsequent cost reduction round of designs offers other EDC vendors a second chance.

OP

NE

XT

Contents Zoom In Zoom Out For navigation instructions please click here

Contents Zoom In Zoom Out For navigation instructions please click here

Next PageSearch Issue

Next PageSearch Issue

All cabinets are not created equal. Only Corning Cable Systems OptiTect Premier Cabinetsoffer the smallest footprint in the industry our latest Local Convergence Cabinets are upto 50% smaller and lighter than previous generations. This can result in up to 40% lowerinstallation cost. Combine that with the freight and storage savings, clear aestheticadvantage and best-in-class fiber management and its clear that no substitute can deliverthis customer value.

Accept No Substitutes.

Often imitated ... never duplicated.The OptiTect Premier Cabinet is an integral part

of the Evolant Solution for FTTx Networks.Visit www.corning.com/nosubs/lw for more

information on Evolant Solutions and receive a free gift.

www.corning.com/cablesystems 800.743.2671 [us and canada] 828.901.5000 [international]

2007Corning Cable Systems, LLC 2007 / EV-01433


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

__________________________________________

1-800-622-77115290 Concourse Drive Roanoke, Virginia 24019

Phone 540-265-0690 www.occfiber.com

No matter where youre located, our fiber optic cable products are there. Optical Cable Corporation has built

a network of reliable stocking distributors and a dedicated sales team committed to getting you the best

fiber optic cables for your requirements when you need them...where you need them. For over 20 years,

we have been manufacturing the broadest range of top-performing tight-buffered fiber optic cables for

most applications in the government, military, and commercial markets. Your order is our top priority.

Contact Optical Cable Corporation for a stocking distributor nearest you. We are where you are.

Our U.S. delivery area.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

Focus on

VO L . 2 4 , N O . 5 M AY 2 0 0 7ContentsTechnology Applications Industry

www.lightwaveonline.com LIGHTWAVE May 2007 5

FOCUS ON 13 TRANSPORT PROTOCOLS Scaling to 100GbE: Drivers

and implementationBY VIJAY VUSIRIKALA, INFINERA

13 TECH TRENDS PON chip space gets

more crowdedBY STEPHEN HARDY

17 STANDARDS WATCH ICEA publishes two fiber-

optic cable standardsBY MICHAEL D. KINARD, CONSULTANT

20 PRODUCT PROFILE Startup makes nanophotonics

device with DUV photolithographyBY MEGHAN FULLER

20 PREMIER PRODUCTS Components, installation

tools, and test equipment

FOCUS ON 23 TRANSPORT PROTOCOLS Multiservice switching platforms

enable global Ethernet servicesBY KEVIN WADE, TURIN NETWORKS

23 CASE BY CASE Verizon Business plots business

plan for 07 and beyondBY MEGHAN FULLER

31 OFC/NFOEC 2007: Recovery underway, but incompleteBY MEGHAN FULLER

31 ANALYST CORNER iSuppli sees promise for

optical in wireline capex forecastBY STEPHEN HARDY

35 PEOPLE Eudyna appoints executive VP

Fluke Networks names president

IEEE presents Photonics Award

38 MARKET WATCH 40G component market

to net $500M by 2012

Worldwide PON equipment neared $1B in 2006

Fiber availability increases, SMEs still underserved

Photonic device takes cue from electronic processes with DUV photolithography page 20

FOCUS ON TRANSPORT PROTOCOLS Our pages are packed with the best approaches to optimizing transport net-works. Some advocate protocol-specific plans for securing packet transport networks, such as OTN and T-MPLS and PBT/PBB-TE (both front page). An-other industry voice says 100GbE services are possible without re-inventing the wheel when it comes to optical infrastructures (page 13). Yet one more advises global carriers to move to multiservice Carrier Ethernet service de-livery platforms to maximize bandwidth efficiency (page 23).

6 update

10 Editorial STEPHEN HARDY Counting chickens

up-front


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

updateHOTPICK

LIGHTWAVE (ISSN 0741-5834), a trademark, 2007, is published 12 issues per year monthly by PennWell Corporation, 1421 South Sheridan Road, Tulsa, OK 74112, telephone 918-835-3161; fax 918-831-9497; Web address www.pennwell.com. All rights reserved. No material may be reprinted. Bulk reprints can be ordered from Kathleen Skelton 603-891-9203 voice, 603-891-0587 fax. SUBSCRIPTIONS: 847-559-7520, 7:30am-6pm CST. Subscrip-tion rates in the U.S. and possessions: one year, $144; Canada/international via surface mail, $163; international via airmail, $197. Periodicals postage paid at Tulsa, OK, and additional mailing offices. We make portions of our subscriber list available to carefully screened companies that offer products and services that may be important for your work. If you do not want to receive those offers and/or information, please let us know by contacting us at List Services, Lightwave, 98 Spit Brook Road, Nashua, NH 03062. POSTMASTER: Send address changes to: LIGHTWAVE, P.O. Box 3279, Northbrook, IL 60065-3279 PRINTED IN THE USA GST NO. 126813153 PUBLICATIONS MAIL AGREEMENT NO. 908584 Return Undeliverable Canadian Addresses to P.O. Box 122, Niagara Falls, ON L2E 6S4

DISCLAIMER: The information contained in this publication is for general information purposes and is not intended to be advice on any particular matter. No subscriber or other reader should act on the basis of any matter contained in this publication without considering appropriate professional advice. PennWell Corporation, and the authors and editors, expressly disclaim any and all liability to any person, whether a purchaser of this publi-cation or not, in respect of anything (and the consequences of anything) done or omitted to be done by any subscriber, reader, or other person in reliance upon the contents of this publication.

6 May 2007 LIGHTWAVE www.lightwaveonline.com

Finisar Corp. (www.finisar.com) has agreed to acquire Wilmington, MA-based AZNA LLC for $19.7 million and South Plainfield, NJ-based Kodeos Communications Inc. for $7 mil-lion. The AZNA acquisition will provide Finisar with chirp-managed directly modulated la-sers (CMLs), which enable longer-reach optical transmitters at a lower cost and with less complexity than those based on externally modulated lasers. Kodeos optical duobinary (ODB) and maximum likelihood sequence estimator (MLSE)-based technologies, mean-while, will give Finisar the 10-Gbit 300-pin MSA transponder product line it is lacking. The acquisitions also bolster the companys expertise in tunable modules.

north americaAlcatel-Lucent (www.alcatel-lucent.com) signed an agreement to ac-quire substantially all assets, including all intellectual property, of Canadian metro WDM networking supplier Tropic Networks Inc. The integration of Tropic Networks intellectual property, namely the Wavelength Tracker, en-hances the competitiveness of Alcatel-Lucents optical product portfolio and gives operators the benefit of cost-effectively upgrading their networks according to the latest innovations in optical transmission, say Alcatel-Lucent representatives. This transaction builds upon a collaboration that the two companies established in July 2004.

Optium Corp. (www.optium.com), a supplier of optical subsystems, has entered into a de-finitive agreement to acquire Kailight Photonics Inc., provider of 40-Gbit/sec optical trans-mission products, for approximately $35 million in cash. Kailights family of optical modules includes 300-pin 40-Gbit/sec transceivers addressing line- and client-side applications. Op-tium says the acquisition will build on its existing suite of optical transport products, includ-ing wavelength-agile DWDM, SONET, and 10-Gigabit Ethernet transceivers and subsystems; 10-Gbit/sec pluggable transceivers; cable TV trunking and distribution subsystems; and its wavelength-selective switch (WSS). In addition, Optium claims the Kailight acquisition will enable it to provide full line cards of all variations for 40-Gbit/sec applications.

Avago Technologies (www.avagotech.com) and Infineon Technologies AG (www.infineon.com) have entered into a definitive agreement under which Avago will acquire Infineons polymer optical fiber (POF) business, based in Regensburg, Germany. The two companies say that Infineons POF business is the leader in the market for automotive multimedia in-fotainment networks and transceivers for safety systems. This business unit also provides transmitters and receivers for transportation switching and home broadband services. All research and development, marketing, and manufacturing employees of Infineons POF group are expected to become Avago employees and will continue to be located at the present Regensburg facility.

AboveNet Inc. (www.abovenet.com) says it has acquired IRUs in, or service agreements for, 60 building access fiber connections held by AT&T in Los Angeles and Chicago as well as 180 building access fiber connections held by Verizon in New York, Washing-ton/Baltimore, and Philadelphia. AboveNet acquired these assets in competitive bid processes that AT&T and Verizon separately conducted to divest certain assets pursuant to consent decrees issued by the U.S. Department of Justice in connection with the mergers of SBC and AT&T, and Verizon and MCI, respectively.

The operating subsidiary of Level3 Communications Inc. (www.level3.com) has purchased certain assets from AT&T Corp.

that were ordered divested as a result of the merger between AT&T and SBC Communications Inc. The assets consist of IRUs for dark fiber connections to more than 200 buildings and more than 1,600 metro fiber route-miles in six of the 11 markets where AT&T was required to divest certain assets. Level3 will acquire fiber assets in Detroit, Hartford, Kansas City, Milwaukee, San Francisco, and St. Louis. Level3 will retain intermediate splice rights, which will enable it to add new buildings to the acquired assets.

OpVista (www.opvista.com) has completed a core network capacity upgrade for Cox Communications throughout the service providers Connecticut and Rhode Island networks. Started in the middle of 2006, the network upgrade features OpVistas recently announced AnyWave Optical Network architec-ture and products, giving Cox Communications a single DWDM platform scal-able to 800 Gbits to support new 10-Gigabit Ethernet services.

Star Telephone, an IOC based in southeastern North Carolina, has selected Ciena Corp.s (www.ciena.com) CN 4200 FlexSelect Advanced Services Platform for its core network infrastructure upgrade. Star is transitioning its legacy SONET network to a next-generation, service-enabled architecture to better support the delivery of high-bandwidth services, including metro Ethernet, wireless, high-speed data, and IPTV.

FPL FiberNet, a subsidiary of FPL Group, has selected Infineras (www.infinera.com) DTN for its regional optical network to bring the benefits of digi-tal optical networking to the growing Florida market, a key connection point between the North American market and the Caribbean and Latin American markets. FPL FiberNet delivers broadband services to IXCs, wireless carri-

ers, international telecom companies, ISPs, and large enterprises. The company has more than 2,500 route-miles of high-count fiber with connectivity to central offices, customer PoPs, carrier hotels, and all interna-tional cable landing stations in South Florida.

CyOptics Inc. (www.cyoptics.com) has announced the signing of a definitive agreement to acquire Apo-gee Photonics (www.apogeephotonics.com), a sup-plier of laser sources for the 10- and 40-Gbit/sec markets. Terms of the transaction have not been disclosed.

March 26, 2007 marked the first meeting of the Serial Short-Reach 40-Gbit/sec Transpon-der (SSR-40) Working Group (www.ssr40.com), established to address the industrys need to provide a cost-effective 40-Gbit/sec serial tran-sponder for short-reach applications. Members include Inphi, Kailight Photonics, Picometrix, Sanmina-SCI, SHF Communication Technolo-gies AG, and Sierra Monolithics.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

___

___

___

3ELECTTHE)DEAL-ULTIPLEXERFOR!LL9OUR,OW*ITTER!PPLICATIONS &UNCTIONAL )NTEGRATED $EVICE $ESCRIPTION &ANOUT /UTPUT/PTIONS 3UPPLY6OLTAGE 0ACKAGE 0RICE

395 $IFFERENTIAL #-,,60%#, 66 XMM-,&

395 $UAL$IFF #-,,60%#, 66 XMM-,&

395, $UAL$IFF (IGH3PEED,6$3 656, XMM-,&

395 $IFFERENTIAL x #-,,60%#, 66 XMM-,&

395, $IFFERENTIAL (IGH3PEED,6$3 656, XMM-,&

395, $IFFERENTIAL x (IGH3PEED,6$3 66 XMM-,&

395 $IFFERENTIAL x #-,,60%#, 66 XMM-,&


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

________________

________________

The Cats Out Of The BagFIS Proudly Announces Our New No-Polish Connector...

161 Clear Road Oriskany, NY 13424Tel: 315.736.2206 Fax: [email protected] www.berinstrumentsales.com

Fastest TerminationLowest Cost

TM

Available in both ST & SC

TM

Starting at$4.95


updateAdesta LLC (www.adestagroup.com), a systems integrator and project manage-ment company for communication net-works and security systems, has been awarded contracts from the Maryland Broadband Cooperative (www.mbdc.us) valued at more than $8 million to design and build the first 140 miles of an open-access telecommunications infrastructure in the states rural com-munities. Adestas turnkey operations will consist of design and construction of the fiber-optic network with 144 to 288 strands of dedicated fiber backbone. Adesta will provide and install an open-access SONET system, with an OC-192 backbone and smaller connections at lat-eral endpoints across segments of the network. These endpoints will connect to universities, industrial parks, and other points of interest, including business de-velopment and technology centers.

Ledcor Technical Services (LTS) and ComSpan-Bandon Networks have se-lected Hitachi Telecom (USA) Inc.s (www.hitachitelecom.com) AMN1220 GPON platform to extend a fiber network that is now operating in Bandon, OR, into

the town of Coquille, about 17 miles in-land from Bandon. Construction is sched-uled for completion in the third quarter. Coquille and Bandon are being connected by fiber-optic cable, which will carry a video feed from the headend, located in Bandon, to Coquille. LTS and ComSpan-Bandon Networks will offer video, high-speed Internet, and voice services.

Corning Cable Systems LLC (www.corning.com/cablesystems) is providing its Evolant Advantage Optical Solution for the Lake Las Vegas fiber-to-the-home (FTTH) deployment. Located 17 miles from the Las Vegas Strip, Lake Las Ve-gas is a 3,600-acre development of residential homes, hotels, and champi-onship golf courses. The development is expected to reach 10,000 units when the build-out is complete.

Connexion Technologies (www.cnxntech.com) has been tapped to build fiber-op-tic networks for three new properties: Highlands at Walnut Creek, a mountain community in western North Carolina (Highlands Mountain Properties); The Park Condos, an upscale mid-rise resi-

dential project in Charlotte, NC (Verna and Associates); and The Point Orlando Resort, a Mediterranean-style high rise in Orlando, FL (Seymour International). The fiber-optic amenity company says it con-tinues to sign additional contracts with national developers at a steady rate.

Santur Corp. (www.santurcorp.com), manufacturer of widely tunable transmit-ters for the telecommunications industry, has entered into a strategic long-term supply agreement with StrataLight (www.stratalight.com), provider of 40-Gbit/sec next-generation optical transponders and subsystems. Santur will supply its pat-ented DFB array technology, which it says enables the fabrication of broadly tun-able sources at the same performance and reliability as fixed DFB lasers.

Xponent Photonics (www.xponentinc.com), manufacturer of surface-mount photonic components, has secured $23 million in a Series 3 funding round led by American River Ventures, which is mak-ing its first investment in the company. Xponents existing investors, including Arcturus Capital, Eastward Capital, El Do-

rado Ventures, HOYA Holdings, Samsung, US Venture Partners, and Walden Inter-national also participated in the funding. According Xponent president and chief executive officer Jeffrey S. Rittichier, the new funding will enable the company to accelerate market penetration of sur-face-mount photonics beyond the FTTX market. He confirms that the company is sampling optical components to the data-com, cable TV, and consumer markets.

emeaBratislava-Orange Slovensko, a sub-sidiary of France Telecom (www.francetelecom.com), has announced plans to build a dense national network of opti-cal connections to households in Slova-kia. Orange believes the deployment will give it a few years head start on its com-petition. The carrier will invest approxi-mately 1 billion SKK (32 million) and cover almost 200,000 households by the end of the year. The optical connections will provide Internet services at speeds of tens of megabits per second and will


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

_

____

___

_________

________

___

___

____________


updatesupport digital television, video- and au-dio-on-demand, telephony, and other ser-vices. Orange says it will deploy optical connections in 10 Slovak cities.

Magyar Telekom, a member of the Deutsche Telekom Group, has signed a multiyear frame contract with Kapsch, a communications service supplier for carriers and service providers, for the supply of Transmode (www.transmode.com) TM-Series CWDM systems. Magyar Telekom says it is using the Transmode TM-Series to extend its existing fiber in-frastructure in metro networks, optimize the aggregation of data traffic in access networks, enable flexible access to high-speed leased-line networks, and provide a cost-effective alternative to dark fiber provisioning. The first deployments under the contract are now live.

T-Com (www.t-com.de), the broadband/landline division of Deutsche Telekom, Germanys incumbent carrier, has se-lected ADVA Optical Networkings (www.advaoptical.com) Fiber Service Platform (FSP) 3000 for the expansion of its opti-cal metro and regional networks. In coop-

eration with ADVA and partner Siemens Communications (www.siemens.com), T-Com signed a multiyear contract for the implementation of WDM-based transmis-sion technology in its next-generation metro and regional networks. The FSP 3000 platforms flexibility and ability to handle multiple network interface proto-cols at data rates ranging from 8 Mbits/sec up to 40 Gbits/sec was reportedly a deciding factor for T-Com.

Vtesse Networks (www.vtesse.com), a U.K. supplier of converged optical net-work services, is implementing BTI Pho-tonic Systems (www.btiphotonics.com) optical networking platform, Netstender, to extend the capacity and reach of its backbone network to support one of the UKs largest financial institutions. Vtesse is leveraging BTIs technology to deliver Gigabit Ethernet connectivity to major fi-nancial sites across a 10-Gbit Ethernet private line backbone. BTIs platform will interconnect networks between Leeds, York, Middleborough, and Newcastle.

Following the commissioning of a trans-oceanic submarine cable upgrade, Azea

Networks (www.azea.net) has secured a Series D funding round of more than $20 million led by TVM Capital and supported by its existing investors. Azea already is backed by top-tier venture capital investors from both the US and the UK, including Ac-cel Partners, Atlas Venture, and Quester. The company reveals that new investor TVM Capital of Munich, Germany, and Bos-ton, MA, has led this additional funding round to provide working capital for further technical and business development and to take the company to profitability.

Enablence Technologies Inc. (www.enablence.com) has agreed to acquire all of the outstanding shares of Albis Optoelectronics AG (www.albisopto.com). The Swiss supplier of photodiodes will continue to supply products to the datacom and FTTH markets as a wholly owned subsidiary of Enablence.

asiaPhotop Technologies (www.photoptech.com) and Wuhan National Laboratory for

Optoelectronics (WNLO) will jointly es-tablish a lab for photonics technologies with applications in fiber-optic communi-cations, lasers, and optoelectronics.

ZTE Corp. (www.zte.com.cn/) has won a contract from MTN, Africas largest mobile carrier, for a Rwandan transmis-sion project to build both the Rwandan backbone network and the first met-ropolitan network in the capital, Kigali. The backbone network project will con-nect Rwandas capital with the Ugandan border, where it will interface with the Uganda MTN network to provide an in-ternational fiber-optic gateway. ZTE will use its ZXMP S330 SDH transmission equipment in building the backbone and metro networks.

Hong Kong Exchanges and Clearing Ltd., one of Asias largest international stock exchanges, has upgraded the speed and reliability of its backbone network infrastructure serving derivatives mar-kets with high-bandwidth optical DWDM technology from Nortel (www.nortel.com). Nortel will provide its Optical Metro 5200 DWDM platform.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

___

___

___

___

___

_________________

Editorial


STEPHEN M. HARDY

Editorial Director & Associate Publisher

[email protected]

My conversation with Steve Rago, principal analyst, IPTV, broadband, and digital home research, iSuppli Corp. not only provided me with the raw material for this months Analyst Corner department (see page 31), it also reminded me of a couple of things.

First is that one of the perks of my job is to be in a position to hear a lot of opinions about a lot of dif-ferent things. Its particularly interesting when I hear conflicting opinionswhich is not a particularly frequent occurrence when it comes to conversations with market researchers. Sure, their numbers never match exactly, but generally speaking, the analyti-cal community tends to be in agreement on major market trends, most recently such things as the rise of Ethernet, the growing market for PON equipment and ROADMs, etc. This isnt surprising when you consider that theyre all covering the same thing and mostly talking to the same people.

Thats why I was interested to hear Rago say that the European market for FTTH, at least in

the near term, would likely be stronger that its North American counter-

part. Ragos statements came just a few days after Vince Chook of Point Topic suggested that the European market was fall-ing behind Asia and the U.S. when it comes to fiber in the access. Pointing to the fact that BTs 21st Century Net-work initiative will focus mainly on xDSL technology

while NTT has already sunk the GNP of a small country into

FTTH, Chook was quoted in a press release as stating, Without

fiber Europe will rapidly become the digital slowcoach on the information

super-highway.Such lamentations about the allegedly sorry state

of European FTTH efforts are pretty typical within Europe, judging by the information flowing through my office in connection with our Lightwave Europe quarterly publication. For example, the majority viewpoint among speakers at the FTTH Council Eu-ropes event in Barcelona this past February appeared to be that while more was happening in Europe this year than last, the carriers on the continent needed to step up their efforts to match those of NTT, KT in Korea, and Verizon in the US. If they didnt, Europe would be put at a competitive disadvantage.

Meanwhile, back here in the States, Ragos opinion echoes that of KMI Research, which estimated last

year that by 2010, Europes Big 5 carriers will have more FTTP subscribers than will U.S. carriers. That will translate into an overall European market for various fla-

vors of FTTX equipment that will be twice the size of the U.S. opportunity, according to KMI.

What are we to make of these conflicts of opinion? Im not sure which is more telling: the fact that there isnt a consensus about U.S. and European FTTH market directions or that the U.S. analysts have Eu-rope in the lead while their Continental counterparts predict the opposite. Perhaps the obstacles to full-scale deployment of FTTH (which differ in some cases between Europe and the US) appear more daunt-ingly large the closer you stand next to them, which makes the prospects on the other side of the Atlantic look more appealing. Or maybe the situation in both markets is in such a state of flux that counting mar-ket chickens before theyve hatched in this instance is more difficult than usual.

Meanwhile, the otherbut relatedthing my con-versation with Rago reminded me of is the danger of extrapolating too far when the analysts do agree. As Ive mentioned, weve all seen the numbers (includ-ing those published in this humble publication) that indicate the growth rates for various segments of the FTTH marketplace. Its easy to get excitedeven when youre an editorby what you read and hear. Thus, it was a useful reality check when Rago said that of the $14 billion that iSuppli predicts will be spent worldwide in access networks this year, only 6% or 7% will go to optical technologyand most of that will go to things like DSLAM backhaul.

Yes, fiber to the home and business should be the future architecture of choice when it comes to the delivery of broadband services. The activity in Asia, within Verizons footprint and among many smaller carriers and municipalities here in the US, and of municipalities, utilities, and telcos in Eu-rope indicate that the right steps are being taken toward that future. But the future isnt here yet. And, as the disagreements among the analysts il-lustrate, not only do we not know when the FTTH future will arrive, were not even sure where it will appear (besides Asia). As always, optimism must remain tempered by realism if the optical commu-nications industry is to maintain a path that is both positive and stable.

EDITORIAL

Editorial Director and Associate Publisher Stephen M. Hardy

603-891-9454 [email protected]

Managing Editor Carrie Meadows


Senior Editor Meghan Fuller


ColumnistsMichael D. Kinard

Editorial Advisory BoardDr. Donald Bossi / Aegis Semiconductor

William J. Cadogan / Vesbridge PartnersAndy Chraplyvy /

Alcatel-LucentDonald T. Gall / Pangrac & Associates

Ira Jacobs / Virginia Polytechnic InstituteMichael Lebby / OIDA

Kevin Lefebvre / SUNY ITStan Lumish / JDSU

Stephen Montgomery / ElectroniCastVik Saxena / Comcast

Paul W. Shumate, Jr. / IEEE LEOS

Presentation Manager Cindy ChamberlinSenior Illustrator Dan Rodd

Marketing Director Lisa A. Bergevin603-891-9410 [email protected] Manager Michelle Blake


ATD PUBLISHING DEPARTMENTS

Art Director Meg FuschettiProduction Director Mari Rodriguez

Senior Vice PresidentGroup Publishing Director

Mark Finkelstein603-891-9133

[email protected]

Group PublisherTim Pritchard


Lightwave Executive & Editorial Offices98 Spit Brook Rd., Nashua, NH 03062-5737

Tel: 603-891-0123, Fax: 603-891-0587

For subscription inquiries only:Tel: 847-559-7520, TDD: 918-831-9566

To subscribe online, go to: www.lw-subscribe.com

In Europe:PennWell European Journal ServiceP.O. Box 4456, London SW12 8AX,

United KingdomPhone: 44 181 675 9775,

Fax: 44 181 675 0301E-mail: [email protected]

CORPORATE OFFICERS

Chairman Frank T. LauingerPresident and CEO Robert F. Biolchini

Chief Financial Officer Mark C. Wilmoth

ADVANCED TECHNOLOGY DIVISION

Vice President of Audience Development Gloria S. Adams

Optimism must

remain tempered by

realism if the opti-

cal communications

industry is to main-

tain a path that

is both positive

and stable.

Counting chickens

In our April Product Profi le (page 20), the interviewees name was misspelled. Dawn Hogh is the vice president of marketing and development at OpVista. Lightwave regrets the error.

Clarification


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

Canobeam Free Space OpticsProvides Maximum Security.

Find out more at canobeam.com

1-800-321-4388 (Canada: 905-795-2012)

Canobeam DT-100 Series, the Free Space Optics system engineered and manufactured by Canon, provides maximumsecurity for wireless transmission. It is a more secure choicethan radio, and engineered to be the best FSO system. Idealfor applications when fiber is not available or too costly to install, Canobeam is the piece-of-mind you need when data

networking, or transmitting voice or video over IP. Plus,because it's manufactured by Canon, you can be confident that support will be there when you need it. Canobeam features built-in AutoTracking in all three models includingtheir best selling Gigabit speed DT-130. For secure wireless network transmissions, trust Canobeam.

How secure is yourwireless network transmission?

It's the choice over radio for high bandwidth (up to Gigabit) transmission.

2007 Canon U.S.A., Inc. All rights reserved. Canon is a registered trademark of Canon Inc. in the United Statesand may also be a registered trademark or trademark in other countries. IMAGEANYWARE is a trademark of Canon.

See us at Interop Booth #2149


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

Want to bring IP to the TV?

Flawlessly?

www.EXFO.com

EXFOs IPTV test equipment provides you with complete solutions, from core to access.Case in point, the handheld CoLT-450P DSL Triple-Play Test Set and the platform-basedFTB-8510B Packet Blazer module. Go to EXFO.com/IPTV, and move your IPTV servicesto the head of the class.

Put us to the test.

From core to access, we test.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

Technology Advances in research, development, engineering, and standards

CONTINUED FROM PAGE 1


TECH TRENDS

Cont. on pg 18Cont. on pg 16

Cont. on pg 14

The sustained high growth of Internet traffic has important scaling implications for IP backbone network capacity, individual core router size, router-to-router link bandwidth, and the opti-cal transport networks used to carry this traffic across the WAN. A number of new applications in a variety of industry verticals are feeding the continued traffic growth in the core. Rapid con-sumer adoption of downloadable video content, facilitated by ultra-broadband access networks, is driving the need for higher speeds in carrier, cable multiple-system operator (MSO), and con-tent delivery networks. And in enterprise net-works and data centers, new trends such as file and storage virtualization, combined with new data sets like high-resolution imagery and video, require higher-speed interfaces.

Scaling techniques currently used to meet such growth typically are based on bundling n 10-Gbit/sec links using Layer 2 link aggrega-tion groups (LAGs) or Layer 3 equal cost mul-tipath forwarding (ECMP). These techniques have a number of drawbacks, including the high cost and complexity associated with multiple ports and links. Moreover, inefficiencies result

from handling high-bandwidth flows because they cannot be striped effectively across multi-ple links. Thus, the limitations of current scal-ing techniques and continued traffic growth are underlining the strong need to develop an eco-system for the next higher speed of Ethernet: 100-Gigabit Ethernet (100GbE).

100GbE WAN transportThe introduction of any new Ethernet standard requires broad industry consensus in definition, implementation, and adoption. Several standards organizations currently are developing different pieces of the end-to-end 100GbE puzzle. The In-stitute for Electrical and Electronics Engineers (IEEE) 802.3 committee, with its Higher Speed Study Group (HSSG), is developing the MAC layer parameters and LAN physical interface specifications for various application distances. The International Telecommunications Union (ITU), meanwhile, is developing recommenda-tions for 100GbE transport over the WAN us-ing the currently defined Optical Transport Network (OTN) rates as well as a new, higher-speed OTN rate. The phys-

By Vijay Vusirikala

Scaling to 100GbE: Drivers and implementation

Connection-oriented packet transportment of PBT and T-MPLS has been driven by a carrier need to migrate from existing SONET/SDH to a fully packet-based network that can easily scale to meet future bandwidth demands. IP-based services are expected to drive these demands. This factor has prompted many car-riers to use IP/MPLS routers as the cornerstone of their migration strategies.

While some carriers have already migrated to a fully IP/MPLS-based network, others, par-ticularly those with large existing SONET/SDH networks, are concerned with the impli-cations of such a step. Since IP/MPLS is based on a network approach different from SONET/SDH, extensive retraining and work-process redesign would be required. There are fur-ther concerns that the complexity involved in configuration and maintenance of IP/MPLS routers will require highly skilled staff. These factors, combined with the cost of router in-

terfaces, have prompted some carriers to seek alternatives.

Carrier Ethernet has gained traction recently as a connectivity method between routers that promises reduced cost and complexity. The widely understood benefits of Ethernet, such as ubiquity, simplicity, and low cost, and the fact that it can be implemented over multiple technologies have driven Carrier Ethernets appeal.

However, today most Carrier Ethernet networks are based on either Ethernet over SONET/SDH or Ethernet over MPLS (using pseudowires/virtual private line services). To date, Ethernet as a standalone transport mech-anism has not been widely considered appeal-ing due to its inability to meet carrier-class requirements.

PBT and T-MPLS provide a new alternative in which the former ad-

Verizons recent rollout of GPON infrastructure, coupled with field trials by Tier 1 carriers in Europe and elsewhere, has firmly established the ITU-T standardized version of PON-based FTTH as a viable market. Chip makers have taken notice; what had once been a niche supported by a pair of vendors suddenly has attracted the attention of a range of companies that are lining up devices based on system-on-a-chip (SoC) approaches. The EPON space hasnt attracted the same number of companies, but the SoC-based com-petition has been equally fierce. In both cases, differentia-tion based on features and cost should prove as important as time-to-market.

PONs comprise an optical line terminal (OLT) in the cen-tral office that feeds multiple CPE devices (called optical network terminals, or ONTs, in the GPON world and opti-cal network units, or ONUs, in EPON) via passive splitters in between. A single line out of an OLT could serve as many as 64 CPEs, although actual numbers will likely be signifi-cantly smaller. Needless to say, that means theres a greater demand for CPE chips than OLT chips.

Thus, its no surprise that there is also a greater number of chip makers chasing CPE than OLTs as wellor at least looking to provide ONT chips before OLT devices. In the GPON space, that means only a few current alternatives to in-house expertise when it comes to OLT development. BroadLight (www.broadlight.com), which consistently has reached the market first with ITU-T compliant silicon, re-cently announced a new quad controller SoC for GPON OLTs, the BL3458, that it says will be available next month. The chip consumes approximately 4 Wless than a watt per line, says Dan Parsons, director of marketing at Broad-Lightand combines integrated burst-mode CDR (BCDR) and SerDes, a dynamic bandwidth allocation (DBA) engine, an integrated processor, upstream and downstream forward error correction (FEC), and downstream encryption and key exchange. Designed to replace four single-line chip com-

binations, the IC supports PON-aware quality of service (QoS).

BroadLight may be the first to de-liver a quad device, but it wont be the last. Iamba Networks (www.iamba.com), which provides technology and IP to companies looking to en-ter the PON market, announced on May 7 a new family of SoCs that are part of its Iamba GPON Eco-System (iGES). The OLT silicon includes the dual-

By Stephen Hardy

PON chip space gets more crowded

Stephen Hardy is the editorial director and associate pub-lisher of Lightwave.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

___

100-Gbit/sec Tx PIC100-Gbit/sec Rx PIC

Source: Innera

100-Gbit/sec Txmodule

100-Gbit/sec Rxmodule

10 10Gtransmitters

10 10Greceivers

10 DWDM demultiplexer10 DWDM multiplexer


Technology


Scaling to 100GbE: Drivers and implementationical interfaces for optoelectronic data modules are typically standardized via multisource agreements (MSAs) among the optical component vendors.

A crucial piece of the 100GbE eco-system is the ability to cost-effectively transport a 100GbE signal across the WAN while easily provisioning, man-aging, troubleshooting, and protecting the service. The 100GbE signal from the client equipment (e.g., router) is connected to the optical transport sys-tem using a short-reach physical in-terface, currently being standardized by the IEEE. This signal is then trans-ported over WDM networks spanning metro, regional, or long-haul distances, after which it is handed off to the re-ceiving client.

The industry has proposed two ap-proaches for this 100GbE over WDM transport: a serial method with a native line rate up to 120 to 130 Gbits/sec and a concatenated approach where the 100 Gbits/sec are inverse multiplexed into multiple wavelengths or optical data units (ODUs) in the OTN framework.

In the serial approach, the 100GbE signal is transmitted on a single wave-length using serial transmission at a rate that is typically higher than 100 Gbits/sec. For example, the serial trans-mission recommendation of the ITU proposes a payload container of 122 Gbits/sec and a transmission rate of 130 Gbits/sec with forward error cor-rection (FEC) overhead.

While the serial transport option looks attractive from an ease-of-man-agement point of view, the technologies needed for serial 120- to 130-Gbit/sec transmission are currently immature and very expensive. Serial transport can be accomplished using TDM-only transmission or by techniques that use advanced signal coding (e.g., duobi-nary transmission) or modulation tech-niques (e.g., QPSK, DQPSK, or QAM) to transmit 100 Gbits/sec at a lower ef-fective baud rate. However, the imple-mentation of both serial transmission techniques requires sophisticated and expensive optical components that are still a few years away from widespread commercial deployment.

Moreover, with these techniques, scaling to 100 Gbits/sec would continue to be limited by a wide range of optical impairments. Many optical transmis-sion impairments, including chromatic dispersion, polarization-mode disper-

sion, and optical receiver sensitivity, scale exponentially with bit rate and limit the unregenerated reach unless complex and costly compensation tech-niques are used. When carriers scale their networks to 40 Gbits/sec today, they face significant issues in terms of compensation for optical impairments. And these impairments are magnified by a factor of 6.25 when scaling the net-work from 40 to 100 Gbits/sec.

In the concatenated approach, the

packet data from the 100 Gbits/sec is inverse multiplexed over multiple ODU channels and deterministically remul-tiplexed back at the receiver. A number of concatenation schemes are possible depending on the type of transport unit chosen. For example, a 100-Gbit/sec signal can be transported using three ODU3 (ODU3-3v), 11 ODU2 (ODU2-11v) or 10 ODU2e (ODU2e-10v). (ODU2 and ODU3 refer to 10- and 40-Gbit/sec rates, respectively, as defined by ITU OTN. ODU2e refers to the overclocked version of ODU2 with a line rate of 11.1 Gbits/sec that some optical transport equipment providers have implemented to support the trans-parent transport of 10GbE LAN PHY.) An implementation based on ODU3-3v results in significant bandwidth ineffi-ciency and does not support compatibil-ity with currently deployed 10-Gbit/sec networks. ODU2e-10v, by contrast, is

attractive because of its seamless com-patibility with the 10GbE LAN PHY transport paradigm. Moreover, the use of an OTN framework for transporting 100GbE signals provides the additional benefit of a digital wrapper for perfor-mance monitoring, fault isolation, and protection of the circuit.

This type of super-wavelength transport through the bonding of mul-tiple 10-Gbit/sec channels over WDM provides a number of benefits compared

with the serial approach. The use of con-catenated 10-Gbit/sec wavelengths en-ables 100GbE signals to be transported using existing optical line systems with-out cumbersome reengineering and ul-traprecise compensation techniques. The rapid adoption of 10-Gbit/sec tech-nologies also has propelled 10-Gbit/sec optical components along a steep vol-ume-price curve, thus enabling low-cost equipment. Finally, the use of concate-nated transport containers enables the transport link to function (with reduced capacity) even under single-wavelength failure conditions.

Integration and super-wavelength servicesThe use of large-scale optoelectronic photonic integrated circuits (PICs) enables efficient, high data rate trans-mission using a super-wavelength approach. A PIC can integrate the

functionality of dozens of optical com-ponents, including lasers, modulators, detectors, attenuators, multiplexers/demultiplexers, and optical amplifiers, into a single device.

The figure depicts a commercially available implementation of PIC tech-nology with a 100-Gbit/sec trans-mit PIC and a 100-Gbit/sec receive PIC, each incorporating multiple op-tical devices onto a chip some 5 mm2. PIC technology may enable signifi-cant improvements in the size, power consumption, reliability, and cost of ul-trahigh-bandwidth optical interfaces at 100 Gbits/sec or higher. PICs operating with 10 wavelengths at 10 Gbits/sec per wavelength for a total DWDM capacity of 100 Gbits/sec have been widely de-ployed. By comparison, each 10-Gbit/sec channel in a conventional system requires up to a half-dozen discrete op-toelectronic components (e.g., lasers, modulators, wavelength lockers, de-tectors, attenuators, WDM multiplex-ers, and demultiplexers).

Large-scale photonic integration may enable even greater capacity and func-tional integration. Recent R&D efforts have demonstrated PICs capable of to-tal aggregate data rates of 400 Gbits/sec and 1.6 Tbits/sec per device pair. The in-tegration and packaging consolidation monolithic integration offers may also enable future optical component costs to follow a cost reduction curve defined by volume manufacturing efficiencies, greater functional integration, and in-creased device density.

Service providers want a cost-ef-fective approach to 100GbE that does not require significantly re-architect-ing their existing transport networks. Serial 100-Gbit/sec transmission may face technical and economic issues that limit its viability for a number of years. Hence, the most viable approach may be to bond multiple wavelengths oper-ating at 10 Gbits/sec into a super-wave-length service. Such super-wavelength services are possible today over opti-cal transport infrastructures flexible enough to carry sub-lambda, lambda, and super-lambda services and allow service providers to deliver next-gener-ation services in a manner that is soft-ware- and protocol-based rather than infrastructure- or network-based.

Vijay Vusirikala is director of techni-cal marketing at Infinera (www.infinera.com). He may be reached via e-mail at [email protected].

Photonic integrated circuits operating with 10 wavelengths at 10 Gbits/sec per wavelength (top) enable transmitter and receiver modules with a total DWDM capac-ity of 100 Gbits/sec (bottom). Each 10-Gbit/sec channel in a conventional system, by contrast, requires up to a half-dozen discrete optoelectronic components.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

__


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

T-MPLStransportnetwork

T-MPLS tunnel

IP/MPLS

PWE3

Backup LSP

Primary LSP

T-MPLSswitch

IP/MPLSrouter

IP/MPLSrouterT-MPLS

switch

Transport MPLS (T-MPLS): Based on ITU-T G.805 architecture Same architecture as SONET/SDH Supports layering Bidirectional LSPs Protected end-to-end Uses OAM for monitoring and 50 msec protection switching

T-MPLS tunnel Supports multiple L2 and L3 services Transports client IP/MPLS LSPs Supports Ethernet and legacy over client pseudowires (PWE3)

T-MPLS multiservice support


Technology


T-MPLS and PBT/PBB-TE offer connection-oriented packet transportdresses the shortcomings of Ethernet while the latter addresses the complex-ity of IP/MPLS. Yet despite their differ-ent starting points, both are strikingly similar and address a key concern of carriers: how to migrate smoothly from an existing SONET/SDH trans-port to a fully packet-based network.

Understanding transport networksTo understand the challenge of mi-grating from SONET/SDH transport networks, one has to understand the nature of transport networks.

Some of the crucial characteristics of transport networks include: Connections are set up infrequently,

but have long holding times. Multiple, often critical, services are

supported by one connection. These aspects require a high level of

protection and availability. Quality-of-service mechanisms are

required to ensure prioritization of critical services.To ensure that these requirements

are met, transport network opera-tors require control and determinism. They want full end-to-end control of packet transport.

These requirements are at odds with the dynamic, autonomous, self-man-agement of connection-less networks such as those based on IP/MPLS. Here, signaling protocols allow the network itself to determine, establish, and re-route paths through the network. Net-work management becomes reactive rather than proactive.

PBT and T-MPLS both provide a packet-based network approach that meets the requirements of transport networks outlined above. The key characteristics that both PBT and T-MPLS provide include: End-to-end, bidirectional point-to-

point connections or tunnels that can support a number of packet-based services.

Protection of these tunnels on an end-to-end basis with the ability to switch over in less than 50 msec.

Assurance of OAM integrity on an end-to-end basis to support protec-tion switching, fault detection, and monitoring of tunnels.

Use of network management systems to control how tunnels are config-ured or provisioned.With PBT and T-MPLS, it is now

possible to envisage a migration from SONET/SDH where the operation of the network does not change dramat-ically, where extensive retraining is not required, and where existing work practices can be retained.

T-MPLS basicsT-MPLS was the first attempt at re-alizing a transport-centric packet network. Standardized by ITU-T (in G.8110.1, G.8112, and G.8121) and based on ITU-Ts G.805 transport net-work architecture, it focuses on sim-plifying MPLS (mainly by eliminating the IP control plane) while adding the OAM and management functionality

needed to meet transport network re-quirements.

The premise of T-MPLS is that MPLS and its associated standards already provide the carrier-class mechanisms and maturity required to successfully implement the re-quirements of a packet transport net-work. The only issue to be addressed is the ability to maintain OAM in-tegrity on an end-to-end basis, by removing IP specific, non-transport related functionality.

Thus, T-MPLS removes the follow-ing functionality, which impedes end-to-end OAM integrity: Penultimate hop popping: This

mechanism pops the MPLS label at the penultimate node, sending IP packets to the last node. This eases the processing at the final node, but

also means that MPLS OAM packets cannot reach this node.

Label switch path (LSP) merging: In fast reroute MPLS link protection, it must be possible to merge two LSPs into one at a node. However, this can create problems in maintaining OAM integrity.

Equal cost multiple path (ECMP): ECMP allows MPLS packets to be sent over multiple LSPs to the same endpoint. However, this makes OAM source identification and processing difficult.By removing such functional-

ity, OAM issues are resolved. The changes allow the use of the follow-

ing transport mechanisms: Establishment of bidirectional, end-

to-end LSPs. Use of ITU-T Y.1711 OAM-based

mechanisms for monitoring and protection switching (Y.17tom and Y.17.tor under study).

Use of ITU-T G.8131/Y.1382 linear protection switching for T-MPLS for 50-msec protection switching (based on Y.1720).

Use of ITU-T G.8132/Y.1383 ring pro-tection switching.Figure 1 shows a typical T-MPLS

network configuration. The T-MPLS network provides a primary and backup LSP where switching can take place within 50 msec. These T-MPLS

tunnels can support both Layer 3 IP/MPLS traffic flows and Layer 2 traffic flows via pseudowires.

The first implementations of T-MPLS concentrate on point-to-point connections configured and mon-itored via a central network man-agement system. However, work is underway to provide automation of configuration using GMPLS as a con-trol plane (similar to existing OTN and SONET/SDH networks).

PBT/PBB-TE operationUntil recently, the Ethernet protocol has not been considered as a viable independent transport mechanism. However, PBT promises to change that perception by addressing the key con-cerns associated with existing Ether-net implementations. These include: The limitations of the VLAN ad-

dress space. The inefficiencies of broadcasting,

flooding, and spanning tree (STP) protocols.

End-to-end monitoring and 50-msec protection switching.The IEEE 802.1ah MAC-in-MAC

frame format is used to provide a very large address space with built-in lay-ering. Building on Q-in-Q and inde-pendent VLAN learning, the customer, provider, and backbone domains are separated with their own addressing spaces (C-VID, S-VID, and B-VID). A new service identifier (I-TAG) is also included to clearly identify the end-to-end transport service. This approach greatly improves the scalability of Eth-ernet networks while enabling protec-tion paths and service transparency.

Traditional Ethernet mechanisms for resolving unknown addresses have been criticized for not being able to scale to the demands of large WANs. Broadcasting and flooding are ex-tremely inefficient on a large scale, while STP does not provide the ro-bust protection switching capabilities required, even in improved versions such as Rapid STP (RSTP).

By turning off broadcasting, flood-ing, and STP, PBT avoids all of these issues. Forwarding tables in switches are populated and controlled via a central management center, but only on the edges of the network and only on initial configuration. Nodes in-ternal to the network can continue to use IEEE 802.1ad provider bridg-ing mechanisms to forward packets, making the use of normal Ethernet switches possible.

Figure 1. Transport MPLS (T-MPLS) provides a primary and backup label switch path where switching can take place in 50 msec.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

PBT end-to-end protection

BackuptunnelB-DA1,B-VID2

Primarytunnel

B-DA1, B-VID1

PBT edge switch

802.1adforwarding

PBT edgeswitch PBT edge

switch

B-DA B-SA B-TAG I-TAG DA SA S-TAG C-TAG Payload data FCS

802.1ad frame

PBT end-to-end protected tunnel: PBT edge switch adds PBT headers Internal switches forward trafc using 802.1ad and IVL Backbone MAC address identies destination PBT edge switch Backbone VLAN ID identies tunnel Combination of B-MAC and B-VID provides working and protected path identiers 802.1ag CFM OAM used for monitoring path and initiating protection switching

PBT uses 802.1ah (PBB) frame format: Backbone address (B-DA, B-SA) Working/protected path identier (B-TAG) Service identier (I-TAG) Includes 802.1ad frame with supplier and customer identiers (S-TAG, C-TAG)

Note: TAG = TPID+VLAN ID

Optional802.1ad

FCS


Technology

The most important requirement, however, is the ability to provide an end-to-end transport path that can be protected and monitored on an end-to-end basis. Since Ethernet is a con-nection-less protocol, this concept has not been addressed to date. By using IEEE 802.1ag Connection Fault Man-agement OAM packets, it is possible at the edges of the PBT tunnel to detect a fault and initiate a switchover to a predefined failover path.

Figure 2 illustrates a typical PBT network configuration, where the pri-mary path is given by the combination of the backbone destination MAC ad-dress (B-DA) and the backbone VLAN ID (B-VID). The backup path has the same B-DA, but a different B-VID to indicate that this is another possible path to the same destination. Thus, multiple paths to the same destination could, in theory, be supported.

With all of these improvements, PBT provides characteristics that make it suitable for use in transport networks. As noted previously, PBT is currently not standardized, but is expected to be part of the IEEEs PBB standardiza-tion under the name PBB-Traffic En-gineering (PBB-TE). It is expected that this work will also consider GMPLS as a possible control plane for configura-tion, as in the T-MPLS case.

Application of PBT and T-MPLSAs can be seen above, T-MPLS and PBT

Figure 2. PBT provides an end-to-end transport path that can be protected and monitored.

are very similar as network concepts, both providing end-to-end, bidirec-tional point-to-point connections that

can be protection switched in 50 msec and centrally managed. With either of these concepts it is now possible to ad-dress migration from SONET/SDH.

Deploying PBT and T-MPLS would keep existing transport network re-sources, work practices, and orga-nizations intact while providing a packet-based network that can scale to meet future bandwidth demand.

Both technologies also provide an al-ternative to router deployment. Rout-ers need only be deployed where Layer 3 capabilities are required with PBT or T-MPLS tunnels providing secure transport of data between routers.

This approach is very similar to existing SONET/SDH and OTN networks and provides an easier transition to packet networks for or-ganizations that are familiar with these transport technologies. In this respect, PBT and T-MPLS are more a replacement for SONET/SDH than a competitor to IP/MPLS, and one can expect to see both connection-less IP/MPLS and connection-oriented packet transportsuch as that which PBT and T-MPLS providecoexist-

ing in carrier networks.Multicast remains an issue still to be

addressed by both PBT and T-MPLS. At this time it is difficult to determine how much of an issue this will be, as current transport mechanisms do not support multicasting, yet support mul-tipoint-to-multipoint client services. IPTV is viewed as the major driver of multicasting requirementsalthough it remains to be seen whether multi-casting will occur in the transport net-work or closer to the last mile.

However, the true test of PBT and T-MPLS is yet to come. Performance in the field with documented savings in capex and opex will be the final de-terminants of the success of these con-cepts. In this regard, it is still far too early to say which will succeed. Yet in one sense they have already succeeded in altering our perceptions of packet-based networking and migration strat-egies, which will ultimately lead to the final solution: fully converged packet-based networks.

Daniel Joseph Barry is director of marketing at TPACK (www.tpack.com).

ICEA publishes two fiber-optic cable standardsBy Michael D. Kinard

The Insulated Cable Engineers Association (ICEA) has published two new optical fiber cable Standards: ANSI/ICEA S-87-640-2006, ICEA Standard for Opti-cal Fiber Outside Plant Communications Cable, and

ICEA S-112-718-2006, ICEA Standard for Optical Fiber Ca-ble for Placement in Sewer En-vironments. The former is a revision of the existing U.S. Na-tional Standard, while the latter is a new subject area.

Significant in this fourth re-vision of ICEA S-640, the gen-eral outside plant cable standard, are the establishment of perfor-mance parameters and require-ments for the L-band at 1,625 nm and the definition of a very low temperature rating for ca-ble. The 1,625-nm requirements are the first definition in the in-dustry of L-band performance for cables. Similarly, the very low temperature rating defines

the North American extreme temperature expectation, 50C, and establishes appropriate performance param-eters at this temperature.

ICEA S-718, the sewer cable standard, is a baseline per-formance standard for cables to be used in storm water and sanitary sewers. It is an outgrowth of and supports the work in this area being carried out by ASTM. S-718 extends the requirements of S-640 with specific consid-erations for cables in the environment found in sewers. New and revised test procedures are included. It also fea-tures an extensive informative annex discussing issues to be considered in sewer applications.

ICEA S-640 has been approved by ANSI as a U.S. Na-tional Standard and is under consideration for adoption into the TIA system. ICEA S-718 is in the ANSI ballot process for approval as a U.S. National Standard. They join three other optical fiber cable standards in the long list of ICEA cable standards. These are: ICEA S-83-596-2001, Standard for Optical Fiber Prem-

ises Distribution Cable. ICEA S-104-696-2001, Standard for Indoor-Outdoor

Optical Fiber Cable. ICEA S-110-717-2003, Standard for Optical Fiber Drop

Cable.

STANDARDS WATCH

Michael D. Kinard is a consul-tant in the communica-tions cable industry. He is vice chair of TIA Subcommittee FO-4.2, Fibers and Cables, and a member of ICEA and the U.S. delegation to IEC SC86A. He may be reached at [email protected].


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE


Technology

TECH TRENDS

CONT. FROM PAGE 13

PON iSL2402 SoC, which is currently available. The iSL2404, a quad-PON SoC, will follow later this year; pin-out and design documentation for

the iSL2404 will be released to part-ners mid-2007. According to Moshe Nattiv, Iambas chief executive officer, the iSL2404 will feature a power-ef-ficient multicore CPU architecture as well as the entire SerDes and BCDR functionality.

Other competition looms on the horizon. Dror Salee, director, strate-gic marketing, Communication Prod-ucts Division at PMC-Sierra (www.pmc-sierra.com), which entered the PON space through its acquisition of Passav, touts the PAS5211, which integrates GPON MAC, SerDes, and management functionality, VLAN bridge support, multilevel frame queuing, Ethernet MAC, encryption, DBA hardware engines, and an em-bedded CPU. Its currently in proto-type as an FPGA.

Meanwhile, Cortina Systems (www.cortina-systems.com) has GPON plans for the assets it acquired from ImmenStar. Prior to the acquisition, ImmenStar had announced the Tu-randot family, which would address GPON OLT and ONT applications. Cortina already had a toehold in the OLT space, thanks to the introduction of a GPON burst-mode SerDes device that vice president of sales and mar-keting Zino Chair says major system houses are currently evaluating. Cor-tina plans to integrate the burst-mode SerDes function into the OLT SoC; the resultant quad device will pro-vide superior density, analog integra-tion, and traffic management features when compared with the competition, Chair believes. Cortina has delivered an FPGA version of the device to lead customers. Like the rest of its GPON

chip line, the company plans to for-mally introduce the OLT product ei-ther late this year or early next year.

Looking for a home at homeWhile theres plenty of elbow room in the central office space, things will get significantly more crowded when it comes to the ONT. BroadLight again achieved first-to-market sta-tus with the BL2338 for single-fam-ily units and the BL2340 for multiple dwelling/tenant units (MDUs/MTUs). The devices feature the Runner mi-crocode-based engine for cell/packet service classification, filtering, reas-sembly, and bridging at speeds neces-sary to support voice over IP (VoIP) and IPTV. They also include a CPU, two Ethernet MACs, a GPON/BPON MAC, SerDes and CDR functionality, and a GEM interface.

BroadLight will have plenty of com-petition. Iamba, for example, has been demonstrating interoperability of its iSN1000 in FPGA form for the last 12 months, according to Nattiv. Iamba plans to release the pin-out and design documentation for the ASSP version this summer. The chip features three main modules: a GPON MAC module, the Iamba Triple Play Traffic Pump (iTTP), and the Iamba Multicore CPU Module (iMCM). The iTTP performs QoS and class-of-service functions including queuing, VLAN support, IGMP forwarding, and frame forward-ing at full wire speed. The iMCM fea-tures several CPUs that handle PON management, voice traffic, DBA, net-work management, and other func-tions. The chip also handles all BCDR and SerDes functions on two Gigabit Ethernet ports to accommodate data handoff and interface to MoCA or HPNA devices.

Meanwhile, Freescale Semicon-ductor (www.freescale.com), Broad-Lights primary competitor in the BPON space, has released the SoC it developed with Alcatel-Lucent, the MSC7120. The chip features a CPU de-signed to deliver 8.5-Gbit/sec peak I/O throughput and a hardwired packet engine for line-rate forwarding of iMix packets, an integrated OMI/CDR, a pair of Ethernet MACs, a GEM MAC, and a 32-bit DDR1/2 memory control-ler. Unlike the BroadLight and Iamba offerings, the MSC7120 also features an integrated DSP for internal support of four channels of VoIP.

While Alcatel-Lucent enjoyed first

crack at the device, other alpha cus-tomers also have received samples of the chip, says Suhail Agwani, market-ing manager for Freescales broadband CPE portfolio. General availability will depend on market dynamics, Ag-wani adds.

PMC-Sierra planted its GPON ONT flag at OFC/NFOEC with a reference design based on the PAS6201-G0. The SoC leverages the design of the com-panys PAS6201 EPON ONU chips. According to Salee, that means use of the companys GigaPASS architec-ture, which uses configurable hard-ware machines that perform data processing to support full 1-Gbit/sec line rates. That enables the con-trol processor to handle other func-tions. The SoC performs QoS queuing, packet classification, VLAN bridging and manipulation, and IPTV filter-ing and features an internal Ethernet MAC. The GPON SerDes and framer functions reside in a companion chip, the PAS6211. Both the PAS6201-G0 and PAS6211 are sampling now.

Cortina Systems, as mentioned pre-viously, matches its OLT work with chips for CPE. The company has both single-family and MDU devices on its roadmap. Traffic management capa-bilities again will prove a significant point of differentiation, Chair believes, as will the devices integrated switch-ing capabilities. Like the OLT offering, formal announcement is expected ei-ther late this year or early in 2008.

Conexant (www.conexant.com) be-lieves it will trump all of these com-panies with the Xenon IIIG CX95202.

The chip, which debuts this month, combines GPON ONT and home gate-way functions into a single device, ac-cording to Rajiv Dighe, director of PON products at Conexant. Full de-tails of the device were not available at press time; however, Dighe did say that the chip will contain a deep clas-sification engine, accommodate four voice channels, and support bridging and multicasting at full line rates. It also will provide complete Layer 3 capabilities supported by an integrated hardware accelerator so you can get Layer 3 performance at hundreds of megabits, while the Layer 2 perfor-mance is at full line rate, according to Dighe. Conexant expects to have sam-ples of the chip in customers hands at the time of its announcement, and hopes to have the device in production either late in the third quarter or early in the fourth quarter of this year.

Several other companies also re-ported having such a device on their product roadmaps; competitors to the CX95202 should emerge quickly.

In the meantime, Dighe says the com-pany also plans to introduce a GPON OLT SoC in the future.

Six GPON ONT sources would seem to be enough to give system design-ers plenty of choice. However, more companies have announced their in-tention to enter the field. Mindspeed Technologies (www.mindspeed.com) announced this past February that it would collaborate with PON systems supplier Terawave Communications (www.terawave.com) to create an ONT SoC for the general market. The chip

PMC-Sierras new PAS6201-G0 gives the company a foothold in the GPON market. The chip is paired with the PAS6211.

PM

C-S

IER

RA

The Teknovus TK3714 ONT chip supports both 1.25- and 2.5-Gbit/sec down-stream traffic so carriers can upgrade their downstream speeds without having to change each ONU.

TEK

NO

VU

S


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

____

__________

____


Technology

will debut later this year; Preet Virk, vice president of marketing for the multiservice access unit at Mindspeed, says the Comcerto 200 will integrate a GPON PMA (SerDes), optical inter-face, and GPON MAC functions with support for DBA, encryption, and FEC. It also will include an approxi-mately 8-Gbit/sec nonblocking Ether-net/GEM switch, two Gigabit Ethernet MACs, and a processor. The SoC will support IPTV HW-IGMPv3 snooping and proxy.

Meanwhile, several sources have their eye on Infineon (www.infineon.com) as a potential competitor. The company currently remains cagey about its plans. As a broadband player, we are heavily interested in the PON business, says Imran Hajimusa, vice president for communications access, North America, at Infineon. There is nothing that has been commercially announced, but we are interestedly watching this market.

With all these companies jumping into the GPON market, it should be noted that at least one has departed. According to Mitch Kahn, AMCCs (www.amcc.com) vice president of marketing for transport products, the company had developed a GPON MAC. But a closer look at the market revealed rapidly declining average selling prices, which made the busi-ness case for fully entering the market questionable, particularly considering the number of competitors. Kahn says AMCC has decided to wait and see if the market evolves in a more attrac-tive way.

EPON marches onCompared with the activity on the GPON end of things, the EPON space appears pretty quiet. Most of the ac-tivity here involves developing devices compatible with the different versions of Asian EPONJapan and China have slightly different standards, with much of the differentiation centered around encryption.

PMC-Sierra/Passav got its start in EPON, where it enjoyed consider-able success. The PAS5201 is its flag-ship EPON OLT device. The chip meets China Telecoms specifica-tions, but can also be used in the Jap-anese and Korean markets, according to Salee. The quad unit integrates an Ethernet MAC, EPON manage-ment, packet classification (includ-ing support for VLANs, video with IP

multicast, IPv4, and IPv6), FEC, en-cryption, programmable DBA, and an embedded CPU. Salee points to the FEC feature as an important dif-ferentiator. PMC-Sierra pairs the de-

vice with the PAS6301 ONU chip, an SoC that integrates MAC functional-ity with SerDes, FEC, Layer 24 clas-sification, and an embedded CPU. The aforementioned PAS6201 preceded the PAS6301 to market; like the PAS5201, the PAS6301 meets China Telecom specifications but can be used in a va-riety of markets.

PMC-Sierras main competition in EPON has come from Teknovus (www.teknovus.com), which also has seen its chips deployed in Japan, Ko-rea, and China. On the OLT side, the single-port TK3721 MAC controller integrates EPON and Gigabit Ether-net MACs and an ARM processor for bandwidth provisioning, among other duties. The chip provides service pol-icy, security, and authentication man-agement as well as DBA control.

The company provides a variety of devices for CPE. The TK3715 com-plies with China Telecoms specifica-tions. It derives from the previously released TK3713, which Teknovus has optimized for IPTV. The TK3713 contains an EPON MAC, SerDes, line-rate Layer 2/3/4 classification and filtering, a switch, bidirectional encryption, FEC, integrated packet buffering, two subscriber ports, and an embedded processor. It supports 40 fully configurable queues (20 in each direction) and VLAN services.

The Teknovus TK3714 ONT chip is unique in that it supports both 1.25- and 2.5-Gbit/sec downstream traffic. Designed for carriers who envision upgrading their downstream speeds without having to change each ONU, the Turbo device has been deployed

in KDDIs network, according to the company.

But the race for design wins in EPON involves more than two com-petitors. ImmenStar had already re-leased a suite of EPON products, the MuLan line, before Cortina Systems acquired it. The chip family, which has been deployed in Japan, included the IS8020 quad-port and IS8030 single-port OLT chips and the IS8010 and IS8015 ONU chips. The IS8010 was optimized for use with ImmenStars OLT chips, while the IS8015 was de-signed to be interoperable with OLT chips from other vendors. Chair says that EPON systems using versions of the chip family for the Chinese mar-ket are currently in certification trials with China Telecom.

Conexant, meanwhile, will comple-ment its GPON ONT/gateway chip with an EPON version, the Xenon IIIE CX95203. (EPON OLT SoCs also are in the companys future.) The device is on the same development schedule as the GPON device. Dighe says versions op-timized for the major Asian markets will be available. The company also

has the CX9520X Xenon II family of EPON ONT chips, which it announced in March 2006. The line supports sin-gle-family, MDU/MTU, and FTTN applications. The ICs are designed to provide VoIP and IP multicasting sup-port, among other functions.

Finally, Centillium Communica-tions Inc. (www.centillium.com) re-cently expanded its EPON offerings to include the Mustang 300 SoC for ONUs. The chip supports traffic man-agement, classification, and filtering for up to 256 multicast groups and up to eight different service types, which Centillium asserts is the highest in the industry. Available now, the Mustang 300 integrates all packet buffer and program memories; the MIPS proces-sor has 192 kbytes of memory, and 1 Mbyte of internal packet buffer mem-ory is included as well.

The Mustang 300 complements Centilliums OLT device, the Colt CE 100. The single-line device integrates an 802.1d bridge, SerDes, CDR, and CPU; it also supports DBA, VLAN, IGMP snooping, and encryption at wire speed.

A single line out of an OLT could serve as many as 64 CPEs, although actual numbers will likely be signi-ficantly smaller.


M SaGEFLIGHTWAVE


M SaGEFLIGHTWAVE

___

____________

_____________


PRODUCT PROFILE

Technology

Though president and chief executive officer Thomas Mossberg calls the half-dozen people who compose startup LightSmyth Technologies (www.LightSmyth.com) a conservative bunch, those same people are responsible for leveraging deep ul-traviolet (DUV) photolithography, borrowed from the electronics world, to develop nanometer-scale optical components. The first such component, a four-channel CWDM multiplexer/demultiplexer, is based on a device that LightSmyth calls a holo-graphic Bragg reflector.

The physical device is similar to an arrayed wave-guide (AWG), says Mossberg, in that it comprises three layers of glass on a silicon substrate, and the light is confined to the center layer. However, Light-Smyth uses DUV photolithography to etch a very fine pattern of lines into the surface of one of those layers, and that pattern serves as a filter, reflecting only light within a certain bandwidth.

For this reason, the LightSmyth device func-tions more like a thin-film filter, even though it is made in the integrated format of an AWG. When light is traveling through the waveguide, its hit-ting the lines that we have patterned and etched us-ing lithographic tools and from each interface, theres a reflection, explains Moss-berg. The sum of all the reflections from those in-terfaces produces the out-put signal. The same is true for a thin-film filter, which is made from multiple thin-film layers, the exact spac-ing and refractive indices of which allow the filter to se-lect specific wavelengths.

While the physics of the thin-film filter and the holographic Bragg reflector may be the same, says Mossberg, the format is radically different. The LightSmyth device is monolithically integrated whereas thin-film filters are intrinsically discrete devices. For this reason, he says you could de-scribe the holographic Bragg reflector as a fully integrated thin-film filterand its not just one filter, he notes. It can be a whole array of them integrated on the same chip.

LightSmyths success hinges on its use of DUV photolithographic patterning tools, which enable quarter-micron and finer resolution. AWGs, by contrast, typically cannot be made much finer than several microns. LightSmyths holographic Bragg reflectors are smaller by a factor of 10.

Moreover, he says, holographic Bragg reflec-tors function via multipath interferencesimi-lar to thin-film filtersand thus provide flexible passband control without adversely affecting the insertion loss. The basic problem with the AWG is that the outputs to go back into the fibers are singlemode, and they only accept a very specific angular range, Mossberg explains. An AWG sends different colors in different angles, but sin-glemode receivers dont like that. Like the thin-film filter, the holographic Bragg reflector sends light back in a single direction. For this reason, he says, we can make very flat-top, controlled pass-band filters with no impact on insertion loss.

Searching for a partnerLightSmyth Technologies has taken its pat-ented holographic Bragg reflector technology and leveraged it in the development of a sin-glemode-compatible, four-channel CWDM multiplexer/demultiplexer. The monolithically integrated silica-on-silicon device is athermal and polarization insensitive to 0.2 dB.

According to Mossberg, development of the CWDM multiplexer is complete, and samples are available; several already have been shipped to select customers. However, the company has decided to partner with a larger pro-ducer to bring the device to market, and Mossberg reports that the company currently is in discussions with potential production partners.

Looking ahead, Moss-berg believes the same de-

vice is compatible with direct implementation in indium phosphide (InP). For now, a key obstacle to silicon photonics is the challenge of coupling in and out of the silicon chips, which is complicated by insertion loss and a high refractive index.

LightSmyth Techologies and others, including Luxtera (www.luxtera.com), are hard at work to overcome this and other challenges.

Mossberg says there is a great deal of con-nection between those areas of interest, of tak-ing silicon photonics and making it practical by eliminating the problem of coupling in and out of the material. For his part, he hopes to see mo-mentum build around DUV photolithography, which he cites as the enabling tool for practical nanophotonics.

By Meghan Fuller

Startup makes nanophotonics device with DUV photolithography

LightSmyth says its four-channel CWDM multiplexer/demultiplexer is the first photon-ic device to take advantage of the deep ul-traviolet photolithographic patterning tools now available in the electronics industry.

Integrated EOE cable assembliesWhile standard copper cable assemblies strain to meet new speed and link length demands in data centers, ZLynx one-piece fiber-optic cable assemblies have inte-grated, industry-standard electrical termina-tions capable of 20-Gbit/sec transmission

in both directions and 5-Gbit/sec duplex four-channel data transmission, for dis-tances up to 100 m. The standard con-nector, compact size, and robust latch mechanism allow ease of use with most equipment chassis. The optical cable uses 3.3 V supplied directly from the electrical port connector. Typical power consumption is just 1 W/connector. Currently sampling, ZLynx products target InfiniBand DDR (dual data rate) at 45 Gbit/sec per channel and 10-Gigabit Ethernet (10GbE) XAUI at 43.125 Gbit/sec applications.Zarlink Semiconductor, www.zarlink.com

980-nm pump technologyThe 980-nm OceanBright pump laser tech-nology is designed to be used in erbium-doped fiber amplifiers (EDFAs). The pump

module incorporates the Bookham genera-tion eight laser chip (G08), is capabl

LW_20070501_May_2007

Documents

Transcript of LW_20070501_May_2007