• TechnologyInnovator
• NextInnovator
• EnterpriseInnovator
• SecurityInnovator
• DefenseInnovator
• WirelessInnovator 
• HPinnovator
• EnergyInnovator
• TransportationInnovator
• SMBinnovator (beta)

• NextInnovator(at)Live.com

• Newsroom Home
• Newsroom Archives
• Fast Facts
• Financial  Information
• Global  Citizenship
• HP Labs
• Company History
• HP in the News   
• Videos
• B-roll
• Blogs
• RSS Feeds 

• Connect
• Connect - Deutschland
• 3000 Newswire Blogs

• Ghost City
• Frontline Sentinel
• Innovation Insights
• WebInno
• Over the River
• Enderle Group
• Security Insights Blog 
• McAfee Audio Parasitics
• Rethinking Security
• Ovum
• iSuppli
• Canalys
• eMarketer 
• CRM Help Desk SW 
• Rethink Research
• The Gadgeteer
• Master the Moment

Today’s new perspective from the WDM & Next Generation Optical Networking Conference (WDM 2012) in Monaco: elastic optical transport is big, the way DWDM was big. As DWDM did, elastic transport promises to address optical fiber bandwidth scarcity. In today’s optical transmission world, a given data rate and bandwidth are used in all optical paths, independent of distance or number of nodes. Elastic optical transmission would allow network operators to trade off distance, spectral bandwidth, and data rate for more efficient transmission. Ciena’s Wavelogic 3 and Alcatel-Lucent’s PSE are the first products that can be used in this way. Other building blocks needed include flexible-grid ROADMs and software-definable transceivers. The relationship between electronics, optics, circuit boards, and modules will be renegotiated. Finally, while not discussed here, a multilayer control plane will also be required to utilize these pieces.

Real deployment should provide new lessons

Matthew Ma of Tata Communications captured my attention with his description of the Ciena gear deployed in Tata Global Network–Atlantic. He described the ability to field-configure the modulation format to maximize capacity per fiber pair based on distance. Wow! He did not describe operational or efficiency benefits, but as these technologies are just being implemented, I’m sure we will learn more with time. OEMs have touted this feature, but the impact was more significant when described from a customer’s point of view. It sounds simple: just better utilization of the available resources. Nonetheless, it is new, not typical of optical communications, and one of many steps needed to realize elastic optical transport.

The technology at the heart of this development is Ciena’s WaveLogic 3. Similar functionality can be realized with Alcatel-Lucent’s Photonic Service Engine (PSE). One of three modulation formats can be selected (BPSK, DP-QPSK, or DP 16QAM). Distance and spectral bandwidth is implicit in the modulation format selection. For example, longer distances can be achieved with BPSK but with a lower data rate, while shorter distances can be covered using DP 16QAM with a higher data rate.

Adaptive modulation format is driven by digital signal process (DSP) technology. The modulation format, symbol rates, and spectral shapes are driven by electronics. The cost of the optics does not change. The implication is lower cost as this technology is amortized over time and deployments.

Very early days of software-definable transceivers

The optical components industry is just beginning to consider transceiver solutions for this application. Deciding where the digital signal processor sits and the nature of the connection to it are fundamental problems. Both Yukiharu Fuse of Fujitsu Components America, Inc. and Tom Williams of Finisar proposed a hot-pluggable transceiver with the DSP on the board. This solution might be acceptable to OEMs with merchant DSP. In this scenario, both Fuse and Williams pointed out, the electrical connector will be challenging. A 32Gbps analog electrical interface is needed, which requires critical and non-trivial development.

Although a pluggable solution for OEMs that do not have captive DSP was not explicitly stated, Williams made a nice argument for the best path forward in this uncertain environment. He advocates designing transceivers that may have unique features but that fit into existing line-card form factors and electrical pin-outs. In this way, different transceivers can be developed and introduced without having to qualify a new line card.

Flexible-grid ROADMs are a must-have; colorless, directionless, and contentionless (CDC) are important too

Flexible-grid ROADMs are clearly a key component in elastic optical networks as variable spectral widths are needed. The role of colorless, directionless, and contentionless ROADMs is less obvious. The value is clearer when considering ROADMs used in conjunction with a control plane where real-time changes are implemented. But as this feels many, many years away, one can question the value of implementing CDC now.

From an operator perspective, however, the easy-to-use, plug-and-play nature of CDC would deliver significant operational benefits from day one, independent of the potential of its dynamic features, making it a must-have for some operators. CDC is a technology that will have a long product ramp cycle because it is expensive, and not all carriers would be willing to pay for it in the early days. The value appears to be there, in conjunction with the other pieces being planned to improve network efficiency.