Reflections and predictions: 2011 & 2012 - Part 1

Gazettabyte has asked industry analysts, CEOs, executives and commentators to reflect on the last year and comment on developments they most anticipate for 2012.

"For 2012, the macroeconomy is likely to dominate any other developments"

Martin Geddes, telecom consultant @martingeddes

Sometimes the important stuff is slow-burning: we're seeing a continued decline in the traditional network equipment providers, and the rise in Genband, Acme, Sonus and Metaswitch in their place. Smaller, leaner, and more used to serving Tier 2 and Tier 3 operators and enterprise players and their lower cost structures. 

The recognition of the decline of SMS and telephony became mainstream in 2011 -- maybe I can close down my Telepocalypse blog as what I foresaw is reality. 

We've seen absolute declines in revenue and usage of telco voice and messaging in leading markets like Norway and Netherlands. The creation of Telefonica Digital is a landmark reorganisation around new markets. No longer are those initiatives endlessly parked in business development whilst marketing dream up a new price plan for minutes, messages and megabytes.

If I had to pick one thing to characterise 2011, it was the year of the App.

For 2012, the macroeconomy is likely to dominate any other developments. The scenarios are "distress", "meltdown" and "collapse".

Telecoms is well-placed to weather the storm. Even £600 smartphones may remain in vogue as people defer purchases like cars and holidays, and hide their fiscal distress with status symbols hewn out of pure blocks of profit. 

Voice will be much more prominent, after decades of languishing, as LTE sets up a complex dynamic of service innovation driven by over-the-top applications - which will increasingly come from telcos as well as telecoms outsiders. Microsoft's purchase of Skype is the one to watch - if they get it right, it joins Windows and Office in the hall of fame; get it wrong, and Microsoft is probably out of the smartphone game due to a lack of competitive differentiation and advantage.

So 2012 is the year when (mobile) voice gets vocal again - because we're going to have a lot to talk about, and want to do it much cheaper and better.

Brandon Collings, CTO for communications and commercial optical products at JDS Uniphase

For the course of 2011, the tunable XFP shipped in volume and it rather quickly supplanted the 300-pin transceiver.  On the service/ market trend, over-the-top consumer video (Netflix) grew rapidly to be the dominant traffic on the internet.

"Solutions for the next generation ROADM networks - self aware networks - are now firm"

I expect the maturation of 100 Gigabit to continue through 2012 with the introduction of a number of new 100 Gigabit solutions, both network equipment makers and at the transceiver level.

Also, as the adoption percentage of consumers using over-the-top video usage still seems to be relatively small, yet is growing strongly and is already the dominant traffic on the internet, it will be interesting to see how this trend continues as it strongly drives bandwidth yet with potentially unfavorable revenue models for the network operators who need to deliver it.  

Lastly, I expect that as the solutions for the next generation ROADM networks - self aware networks - are now firm, the practical assessment of the value and advantages of these networks can quantitatively take place.

Eve Griliches, managing partner, ACG Research @EveGr

The Juniper PTX announcement really caught the market by surprise. I'm not so much sure why but clearly it rocked some folks back on their heels. Momentum for the product has been good as well. I think you can count this as a success story.  

Another one is the Infinera 500Gbps release with super-channels.  A pretty impressive technology and service providers are waiting for final product to test.   

The death of Steve Jobs rattled us all. I think it struck a note for everyone in how different he was and how he touched us all.  

"Content providers ask for simple, scalable and low-featured products. Those who deliver will be rewarded for listening."

I continue to be amazed at how much optical equipment content providers [the Googles, Facebooks, MSNs of this world] are deploying and how few folks at the vendor level are doing anything about getting into their networks.  Maybe that is a 2012 thing, I don't know.

As for 2012, we'll definitely see some mergers and acquisitions - expect low acquisition prices too - and some companies exiting this market.  I love optics and it really pains me to say that, but there are just more companies out there who can't support the declining margins. I think margin erosion will be key to who survives.  

Cisco and Infinera should be bringing some cool products to market in the next six months. We hope the products are good because it will generate debate for the final vendor choices for operators such as AT&T and Verizon. 

Again, content providers ask for simple, scalable and low-featured products. Those who deliver will be rewarded for listening. Some don't listen, and will wonder what happened.

Peter Jarich, service director, service provider infrastructure, mobile ecosystem, Current Analysis @pnjarich

2012 is going to be the year for LTE-Advanced (LTE-A). Why? One, vendors always like to talk up what’s next, and LTE-A is what follows LTE (Long Term Evolution). 

At the same time, operators who haven’t yet deployed LTE will want to look to start with the latest and greatest. Of course, LTE-A brings real advances for operators: carrier aggregation for dealing with fragmented spectrum assets; heterogeneous networks for dealing with the interaction of small cell and macrocell networks; relaying for improved cell edge performance.

Avi Shabtai, CEO of MultiPhy

The most significant development of 2011 was the availability of CMOS technology that allows next-generation optical transport solutions for 100 Gigabit. And specifically, metro-focused solutions that hit the cost and power numbers required by this industry.

On top of that, optical communication has entered the era of digital signal processing receivers. We have also seen the potential segmentation in 100 Gigabit of metro versus long-haul, each with its specific set of solutions.

"We will see a huge growth in video consumption. This has already started but it is just the tip of the iceberg."

The transition of the telecom and datacom market to 100 Gigabit has also begun - from the transport optical network all the way to copper backplanes - it's all a 4x25Gbps architecture. This year has also seen consolidation in the ecosystem, especially among module companies.

This consolidation will continue at all industry levels in 2012: semiconductors, subsystems, systems and the carriers. The consolidation will coincide with an across-the-board price reduction in emerging technologies like 100 Gigabit transport.

The increase in capacity demand will also force an increase in requirements for various solutions supporting 100 Gigabit. I expect to see more CMOS-based devices introduced.

From a services point or view, we will see a huge growth in video consumption. This has already started but it is just the tip of the iceberg. Video will have a tremendous influence on network evolution.

Gilles Garcia, director, wired communication at Xilinx @gllsgarcia

The CFP2 and CFP4 optical modules are arriving a lot faster than it took for the CFP to follow the XFP optical module. 

The CFP standard took 3-4 years to complete while the standard for the CFP2 just closed after two years. Now the CFP4 standard has been launched and is expected to take 18 months only. The new form factors are being driving by the cost-per-port of 100 Gigabit and how to reduce it. The CFP2 doubles the density when compared to the CFP while the CFP4 doubles it again.

"Programmability is becoming the key trend among telecom system vendors as operators look to react faster to standards, new feature requests and deployment of new services."

Telecom application-specific standard product (ASSP) players have been relatively quiet in 2011. Word from customers is that such vendors are pushing out their roadmap/ product availability because of too much flux in the various IEEE and ITU-T telecom standards and difficulties to justify the return-on-investment. This is proving a perfect opportunity for FPGAs.

Large system vendors are growing their network services as operators continue to outsource their network management and maintenance. As reported in their financial reports, this is an important source of business for the likes of Ericsson, Huawei and Alcatel-Lucent. 

It is leading the vendors to push more of their own hardware, as they look to add value-add services and integrate the services using their own platforms. Some equipment vendors realise they do not have a full portfolio and have established partnerships for the missing platforms. They are also starting to develop platforms to generate more revenue.

In 2012, I’m not expecting a telecom revolution but I do expect accelerated evolution. And I foresee big disruptions in the ASSP market as it continues to consolidate: I expect several mergers and acquisitions among the top 20 ASSP suppliers.

Programmability is becoming the key trend among telecom system vendors as operators look to react faster to standards, new feature requests and deployment of new services. Programmability also improves time-to-market to deliver these services and reduce time-to-revenue.  

Mobile backhaul will be a market driver in 2012. The growth in mobile data terminals will lead to a new generation of mobile backhaul networks. This will drive the move from 1 to 10 Gigabit Ethernet, higher-feature packet processing, and traffic management integration into mobile infrastructure to better control and bill bandwidth usage i.e. pay for what you use.

The 'God box' - packet optical transport systems and the like - are back, but really it is network needs that is driving this.

And one topic to watch that will become clearer in 2012 is how cloud computing impacts the networking market with regard such issues as security, cacheing and higher speed links.

Google is becoming an important internal - for its own usage -networking equipment player. And Google will be joined by others - Facebook, Amazon etc.  What impact will this have on the traditional system networking vendors? Such new players are defining and building networks platforms tailored for their needs. This is competition to the traditional system vendors who are not getting this piece of the business. Semiconductors, including FPGAs, could serve those companies directly.

Other issues to note: What will Intel do in the networking space? Intel acquired Fulcrum in 2011 and has invested in several networking companies.

There are also technology issues.

What will happen to ternary content addressable memory (TCAM)? Broadcom's acquisition of NetLogic Microsystems has created a hole in the TCAM market. Will Broadcom continue with TCAM? Will customers want to give their TCAM business to Broadcom?

Xilinx FPGAs have added network search engines IP in the solution portfolio as multi-core ‘search engine’ face increasing difficulty in sustaining the performance required. 

And of course there is the continual issue of power optimisation.

For Part 2, click here

For Part 3, click here