LightTrends

 

ARE NETWORK OPERATORS OVERINVESTING IN 100G NETWORKS?

The short answer is yes, and they should continue to overinvest as much as they can, while the global economy is growing.

Read this article in Chinese

Deployments of 100G optical transport equipment in long-haul networks has been a great success story so far. According to our quarterly sales database released along with our Market Update Report, shipments of 100G DWDM ports reached 15,000 in first half of 2013 and are projected to climb to 40,000 for the full year. While these numbers remain modest compared to shipments of 10G and even 40G DWDM ports so far, the contribution of 100G ports to network bandwidth is already significant.

Data presented in the figure below compares the bandwidth growth rate of global DWDM networks with the growth rate of internet traffic. In an ideal scenario, the two growth rates should be well correlated. However, the reality is strikingly different:

  • Growth in network bandwidth was off the chart during the telecom bubble of 1999-2001.
  • Investment into telecom infrastructure remained depressed in 2002-2005 and the growth in network bandwidth was very low.
  • Correlation between internet traffic and network bandwidth growth rates improved in 2007-2008, but the financial crisis of 2008-2009 led to a decline in network upgrades.

chart

Our analysis of port shipments indicates that network bandwidth is growing faster than internet traffic in 2012-2013 and a significant fraction of this growth is driven by deployments of 100G ports. The number of 100G DWDM ports is expected to almost triple in 2013. Will it double again in 2014? How fast will the 100G port shipments grow in 2015-2018?

 “An ideal world scenario”

Let us assume that network bandwidth and internet traffic were perfectly balanced by the end of 2008, meaning that network operators had a comfortable reserve of bandwidth to handle fluctuations in the internet traffic. Data presented in the figure above suggests that network operators were underinvesting during the economic crisis of 2009 and the slow recovery of 2010-2011. An increased level of investments in 2012-2013 seems to be well justified. In fact, in order to catch up on underinvestment in 2009-2011, network operators will have to keep the rate of network bandwidth growth (solid red line) above the traffic growth (blue line) until 2018, when the two curves merge. This is a purely mathematical modeling and it is very unlikely to happen, given how rocky the past data looks. 

Will the history repeat itself?

If it does, we should expect another downturn. Where the crisis will come from and how it will look like is anybody’s guess. Given the increasing influence of China on the world economy, one scenario leading to a slowdown in 2016, is related to the completion of the current 5-year economic plan in that country. Many projects in China many be accelerated in 2014-2015 to meet the plan’s targets and 2016 may be a perfect time to take a pause.

Regardless of the origin for the next economic crisis, the ability of network operators to finance infrastructure investment will become limited and they should continue to overinvest, while the global economy is growing.

LightCounting Forecast (not shown in the figure)

Market Forecast Report released by LightCounting in July 2013 relies on a more refined model than the scenarios described above, but it is also based on correlation of bandwidth growth and internet traffic. One of our key assumption is that network bandwidth grows slower than internet traffic, as network operators use the available bandwidth more efficiently, limiting the long-term outlook for 100G deployments in core networks. A short-term concern is the very large number of 100G trials and early installations reported in 2013, which may not lead to a sustainable demand for 100G ports in 2014-2015.

For example, some operators may install a few 100G ports now just to make sure that their network is compatible with 100G technology and they can count on it for future upgrades. Also, smaller vendors like Infinera were extremely aggressive in chasing early adopters of 100G in 2012-2013. Will they be able to maintain the same growth rate in 2014? The last earnings report from Infinera does not offer much comfort on the near-term outlook.

Finally: how realistic are expectations that lower cost 100G optics will start selling into the higher volume metro and much higher volume datacom markets? Our expectations for 100G in data centers for the next 2-3 years are modest and we have just reported on the dominance of 10GigE and 40GigE in 2013. The loosely defined metro market can be pulled on either side of the argument. Financial companies and Internet exchanges are deploying 100G in metro now, but how many other metro networks really need, and are ready to pay for massive installations of 100G ports today? Emerging islands of 100G in the metro may not raise volumes high enough for the cost of the 100G components to come down.  

Upcoming Event Participation:

The LightCounting team will be at the following industry events:

  • IEEE Components, Packaging and Manufacturing Technology Society, November 9, 2013
  • SC13, Colorado Convention Center, Denver, November 19-21, 2013


Latest Research from LightCounting:

About LightCounting Market Research

LightCounting is a leading optical communications market research company, offering semiannual market updates, forecasts, and state-of-the-industry reports based on analysis of primary research with dozens of leading module, component, and system vendors as well as service providers and other users. LightCounting is the optical communications market’s source for accurate, detailed, and relevant information necessary for doing business in today’s highly competitive environment. Privately held, LightCounting is headquartered in Eugene, Oregon. For more information, go to http://www.LightCounting.com, or follow us on Twitter at http://www.twitter.com/lightcounting.