Gazettabyte

Gazettabyte asked industry figures for their views after attending the recent OFC/NFOEC show held in Anaheim. Part 2

Bill Gartner, vice president and general manager of high-end routing and optical business unit at Cisco Systems.

There were several key themes during this year’s OFC conference, but what I found most compelling were the disruptive trends and technologies that stand to significantly impact the optical communications market in the coming years.

"SDN could be the single biggest disruptor in the transport industry and has the potential to transform network programmability and orchestration"

One of the hottest themes at this year’s OFC conference is the role of silicon photonics and the benefits it presents to service providers and carriers. Silicon photonics is truly one of the most interesting advancements taking place in the industry as it has the potential to drastically lower the density, power and overall cost of ASICs.

Several carriers at the show, including CenturyLink and AT&T, presented their view that optics is becoming a larger portion of their spend and now exceeds the cost of packet switching technologies.

A second key trend coming out of the show is software-defined networking (SDN) and its impact on networking. There is tremendous industry interest around this topic and it extended to the Anaheim Convention Center.

With SDN, our customers can increase flexibility in terms of selecting the features and protocols that make sense for their network application – whether it is a data centre application, a service provider application or a large-scale enterprise application.

The last theme that resonated during OFC was around the convergence of packet and optical solutions. As service providers look for ways to decrease both CapEx and OpEx related to the network, incremental technology improvements will decrease costs. However, for many customers, their network capacity is growing far faster than their revenues, so incremental improvements will not yield required reductions.

"As an industry we have to evolve organisationally and technically. Those who fail to recognise that face extinction."

This shows us that we need to explore more fundamental shifts in architectures that have the potential to yield significant savings in OpEx and CapEx. Enter the convergence of IP and optical – this may take the form of converged platforms, but will also involve multi-layer control planes that allow the exchange of information between the packet and optical layers. This convergence helps answer questions like: How well is the network utilised? Can it be optimised? Are there multi-layer protection/ restoration schemes that make better use of the available resources?

During the conference, I had the opportunity to present at the OSA Executive Forum, which brought together more than 150 senior-level executives to discuss key themes, opportunities and challenges facing the next generation in optical communications.

What struck me is that this industry is constantly evolving, which presents challenges and opportunities. We are looking at an industry that is highly fragmented at the moment and requires further streamlining.

You have new players at every level of the value chain that bring exciting, unique perspectives and advanced technologies that increase efficiency and decrease costs. But none of this innovation comes without change; as an industry we have to evolve organisationally and technically. Those who fail to recognise that face extinction.

"This is like solving a simultaneous equation where the variables are power, cost and density – you need to solve for all three"

The key themes discussed at OFC are an indication of what is to come in optical transport and mirror our top priorities at Cisco.

In the coming year, we expect to see CMOS photonics technology enable lower power pluggables. This is the case with CPAK, but more broadly, we will see this technology find its way into low cost board-to-board interconnect and chassis-to-chassis interconnect.

As an industry, we have made great progress in reducing the cost of transmitting bits over a long distance but much more remains to be done. As bit rates increase to beyond 100 Gigabit, we must look for ways to drive this cost down faster, while decreasing both power and size. This is like solving a simultaneous equation where the variables are power, cost and density – you need to solve for all three.

During the next five years, I think that SDN could be the single biggest disruptor in the transport industry and has the potential to transform network programmability and orchestration.

We will see an entire software industry emerge around SDN, but it is important to note that this is really all about multilayer control – Layer 0 to Layer 3. SDN is not simply an optical transport problem to be solved. The advantage will go to those who are looking at this holistically.

Brandon Collings, CTO of the communications and commercial optical products group at JDSU

I found it interesting that the major network equipment manufactures had a significantly increased presence on the exhibition floor.

"This year’s focus and buzz was all on silicon photonics with researchers leveraging it against nearly every function in telecom and datacom"

I learned a lot about SDN at levels above the photonic network. This is a very complex topic likely to take some time to fully mature within telecom networks; however, the potential values appear compelling.

This year’s focus and buzz was all on silicon photonics with researchers leveraging it against nearly every function in telecom and datacom. I expect it will be interesting for industry watchers how this promising technology evolves within the industry, where it achieves its promise and where it runs into practical roadblocks.

Vladimir Kozlov, CEO of LightCounting

This was the best OFC since 2000. The optical community is once again energised. Some attribute the improved mood to high-value acquisitions of companies LightWire and Nicira that were made last year, but this is just part of the story.

Yes, the potential of silicon photonics and software-defined networking (which LightWire and Nicira were focussed on, respectively) do broaden the horizon for optical technologies in communication networks and data centres. But the excitement is not limited to just these two ideas. All the new - and old or forgotten - ideas, technologies and products once again have a shot at making a difference. Demand for optics is strong and the customers are hungry for innovation.

"Demand for optics is strong and the customers are hungry for innovation"

In contrast to 2000, few people are getting carried away with the excitement. The mood is much more constructive this time and it makes me hope that most of this new energy will not be wasted.

I would not single out a specific technology or application to watch out for in the next few years. All of them have opportunities and challenges ahead. We will keep track of as many developments as we can and make sure that hype does lead the industry off the tracks this time.

Effie Favreau, marketing, Sumitomo Electric

One hundred Gigabit technology is here. Last year there was a lot of hype about 100 Gigabit and now it is reality; vendors have products that are shipping.

Sumitomo and ClariPhy partnered on pluggable coherent modules. Together, we hosted an impressive demonstration with all the components to make pluggable coherent modules available next year.

"For the enterprise/ data centre, vendors requiring low cost, high density equipment really need the CFP4"

One thing I learned from the show is that vendors need to re-purpose their existing equipment. There was much discussion regarding software-enabled applications and passives to enhance the performance of networks and make them more intelligent.

There was the introduction of the CFP2 from several vendors as well as Cisco's CPAK. For the enterprise/ data centre, vendors requiring low cost, high density equipment really need the CFP4. At Sumitomo, we are concentrating our R&D efforts on the CFP4.