Many softswitch vendors will remember 2005 as a good year. Not a great year, but a good one for getting back on track. From a revenue perspective, it was less than experts had predicted back in the day, but from a foundation-for-the-future perspective, it was more than enough to ease the minds of slump-weary vendors. Now they just have to get used to the notion that there will be no breakout year for the softswitch market. They will have to become familiar with a slower but steadier pace of progress.

In 2001, Dittberner Associates said softswitch sales worldwide would reach $2.8 billion by 2010. At approximately $800 million this year, the pace will have to more than triple over the next four years to reach that goal. It could happen.

Everyone knows what derailed the telecom market shortly after those predictions were made. Since then, there has been a lot of re-positioning, re-thinking and rejuvenation. And as Tom Valovic, program director of voice-over-IP infrastructure at analyst firm IDC, said, “The major paradigm shift has finally taken place.”

He says that's another way of saying that next-generation network (NGN) systems are now really the first choice for a lot of Tier 1 carriers in terms of network planning. Unfortunately for anyone looking for a boom year in 2006, planning doesn't necessarily translate directly to implementation. The shift has occurred in the thinking, not necessarily the doing.

With the paradigm shift in thinking, all the major questions have been answered, doubt about the proper architecture has been removed, and the technology and its vendors have matured. Most everyone now sees a way forward. And that way is called IMS, or IP multimedia subsystem architecture.

“IMS was the biggest development for softswitching in 2005, without a doubt,” Valovic said.

He said that IMS picks up where NGN left off and, in fact, resurrects it. “We see IMS as a refinement of the NGN evolution, not a completely different animal,” Valovic said. “The original NGN wave only achieved limited success in terms of a multi-vendor environment, but if you look at what SBC Communications is planning to do with respect to IMS, you really see the multi-vendor phenomenon taking off.”

By virtue of its standard and distributed architecture, IMS defines the way the softswitch will be deployed. It is one of several developments that position the softswitch market for a smooth migration path forward.

“This whole notion of finally disaggregating the service layer and infrastructure layer is probably the most important trend,” Valovic said.

Another positive trend, probably only in hindsight, is the softswitch market itself. During the economic hiccup, vendors have consolidated, weeded out the weak and enhanced their portfolios in anticipation of both a turnaround and IMS.

“The Tier 2 market is pretty firmly established now, and the start-up market is nowhere to be found,” Valovic said.

He added that Tekelec's acquisition strategy has helped it emerge as a major Tier 2 player and that Sonus Networks has come through difficult financial problems and landed in a good position.

Start-ups such as Cirpack, Santera Systems, Taqua Systems and Telica have been acquired, and small providers such as CopperCom, MetaSwitch and Veraz Networks, as well as Lucent Technology's Telica switches, have staked their claim in the Tier 2 and Tier 3 markets. And some pioneers in the market, such as VocalTec, have grown quiet, at least until its merger agreement with VoIP gateway provider Tdsoft Communications late last month.

Also, the leader board has remained pretty stable over the last several quarters, with the exception of Siemens making a charge on the back of Cablevision, which is beginning to sign voice subscribers at a good clip. Alcatel, Cisco Systems, Huawei Technologies, Lucent and Nortel Networks are holding their own.

Consolidation may not be over, but regardless, if the remaining vendors can deal with the gradual switch to next-generation technology, then there will be a big enough pie for them all to share.

However, the pie will be served in stingy bites rather than the feast that the market once anticipated. Why? Primarily because circuit switches have not yet outlived their usefulness. Nor has the business case been made for not squeezing as much investment as possible out of them. Valovic said that a couple of years ago, everyone was talking about massive infrastructure conversion involving Class 4 and Class 5 replacement.

“A lot of vendors have backed away from that now,” Valovic said.

They have done more than that. Together with their customers in nearly all markets, legacy switch vendors are still growing their circuit-switched infrastructures. Lucent, for example, is still shipping Class 5 equipment.

“We are still shipping both new and growth equipment on our Class 5 switches to customers in all our global markets,” said Jim Kukla, director of Lucent's converged core solutions group.

And although he admits they are fewer now, Kukla said Lucent continues to put out feature releases in the circuit switches that implement product enhancements, have value and are requested by its customers.

Siemens also is still shipping switches outside of Western Europe to markets in Russia and South America.

It is hard to imagine any massive infrastructure conversion with investment, albeit comparatively small, still being made in circuit switches.

And although Lucent's Tracy Coutre, director of Lucent Worldwide Services, said the company hasn't been asked to rework maintenance contracts on their legacy switches — because “[customers] are still using the large embedded base of lines and trunks that have already been deployed” — the practice has been widespread in Europe and will likely extend the life of those switches.

“Established PTTs in Western Europe spent millions of euros back in 2002 renegotiating maintenance contracts and doing some investing in their Class 4 and Class 5 networks,” said Lilian Tau, CEO and executive vice president of market research and consulting at Dittberner Associates. “So they can hardly justify spending more money right now.”

Tau referred to projects such as BT's 21CN projects as isolated, although the outlook for next year is good.

“We expect some of the established entities within Western Europe, like France Telecom and Deutsche Telekom, to become more vested in 2006. In fact, we know it will happen because we are working with them,” Tau said.

Although the numbers aren't in yet for the third and fourth quarters of this year, Tau expects them to be similar to the first two quarters, which were $400 million combined. Although Dittberner raised its 2001 projection of a $2.8 billion market by 2010, Tau doesn't see the market becoming huge until it reaches the point where telcos must replace existing infrastructure.

Given that the U.S. was among the first to embrace digital switching, the time has come, Tau said. “They have to start getting rid of them.”

The age of its switching infrastructure, not necessarily the desire to rush into the next generation, is what prompted Verizon to be a little more aggressive than SBC or BellSouth at switch replacement, Tau said.

A report by Dittberner from September, called “Worldwide NGN Migration Status and Vendor's Opportunity Analysis,” showed that although the U.S. and China lead the world in the number of next-generation ports installed with 23.3 million and 24.7 million, respectively, they also far outnumber other countries in the amount of legacy digital switching ports. China has 782 million, the U.S. has 765 million.

That puts the U.S. at 14th among the top 20 countries in the ratio of digital switching ports versus NGN ports with a 3.05% ratio. The U.K. is highest among large countries at 17.38%. However, in the U.K., a single carrier, in this case BT, accounts for 85% of the installed NGN base.

Dittberner said that the top 20 countries account for 94% of all NGN ports and that Class 4 (48.4%) and Class 5 (30.7%) replacement dominates as a driver when compared with other drivers such as long-distance bypass, offload, cable IP and voice-over DSL.

But perhaps most telling in terms of opportunity for softswitch vendors is the age factor — how long existing digital switches have been in place.

Given the low percentage of new digital switch ports in countries such as Finland, Israel, Italy, Japan, New Zealand, the U.S. and Venezuela and the corresponding high number of switches in those countries that are at least five to 10 years or older, these markets are prime candidates for accelerated switch replacement (see figure).

Eighty-three percent of the digital switches in the U.S. are between five to 10 years old. If they are at the high end, then the next few years will be decision time for many operators. Still, 18% of digital switches are more than 10 years old.

Much of that replacement, however, will occur as carriers merge, which reduces the actual number of new switches. Still, in addition to the raw numbers on the age of its infrastructure, a couple of other factors indicate that the U.S. is ripe for softswitch deployments, according to Dittberner:

• The sizable cable TV infrastructure supports 80 million subscribers, and so far only 27% have been converted to IP with cable modems.

• The competitive environment is healthy, contrary to popular opinion, and accounts for half of the installed NGN ports.

• The U.S. has a very strong environment for fiber-to-the-x (FTTx) development, arguably the best in the world, and the environment for 3G expansion also is favorable because of the number of operators (15) and wealthy consumers.

Dittberner ranks Hong Kong (China), Ireland, Italy, Japan, Sweden and the U.S. as countries with the highest potential for FTTx deployment.

Both Valovic and Tau agree that the immediate growth in the softswitch market will come from wireless operators inspired by IMS. But if current trends hold in regards to which vendor is doing business in which countries, it won't be hard to figure out what companies will get the lion's share of certain business, unless more crossover occurs.

Only Nortel, Ericsson and Siemens — and to some extent Sonus — can boast of spreading their business around to more than two countries. Of the vendors with respectable market share in other countries, 44% of Nortel's NGN business is in the U.S., 73.5% of Ericsson's is in the U.K., and 62.6% of Sonus' and 26.7% of Siemens' NGN business is in the U.S. However, Nortel does 15% in China, 7% in Canada and 8.6% in the U.K. Siemens is the most evenly split with 13% going to Canada, 14.4% to Germany and a little more than 5% to both China and Sweden.

On the other hand, companies such as Huawei (92.8%), Lucent (80.4%), Tekelec (98%) and UTStarcom (93%) do the majority of their NGN business in their home country.

While vendors wait for service providers to retire their legacy switches, there is another segment that falls under the distributed IMS model that is moving at a faster pace and offers growth opportunity: application servers. However, here, the leaders in next-gen switching have been taking a backseat to smaller application server vendors. BroadSoft leads this space with 18.86% of the market followed by NetCentrex (13.79%), Sylantro (10.31%) and Cisco (9.75%).

Nortel, Ericsson, Tekelec, Siemens, Alcatel and Sonus have lagged behind as they range from 8% to below 1% of the market as of the second quarter of 2005. This disparity identifies either potential acquisition opportunities or the large vendors waiting for the green flag to signal the start of the race. Tekelec already made its move last year with the acquisition of VocalData.

In the two years the market will have to wait, according to Dittberner, before the trickle of digital switch replacements turns to a steady stream, we could see a lot of activity in this application server space. Perhaps then, in 2007 or 2008, we see a vintage year for softswitching.

AGE OF SWITCHING

The following table summarizes the changes in digital switching (DS) ports installed over the past 10 years as a percentage of the current installed base (end 2004). For example, if the percentage change during the last four years is greater than 50%, the rating is minimum because this indicates that a large portion of the DS infrastructure is recent. Conversely, if it is below 10% and the percentage of ports installed more than 10 years ago is superior to 50 %, the rating is the highest (Finland, France).

Among the top candidates for digital switch replacement based on age of existing infrastructure

Country	> 10 year	5-10 years	<4 years
New Zealand	55.60%	45.30%	-0.80%
Bermuda	40%	59.60%	0.10%
Finland	68.30%	36.30%	-4.70%
France	55%	44%	1%
Germany	35.20%	61.20%	3.60%
Israel	22.40%	80.80%	-3.20%
Italy	41.50%	54.60%	3.90%
Japan	42.40%	59.30%	-1.70%
Norway	30.40%	72.90%	-3.30%
Puerto Rico	58.80%	39.20%	2%
United States	18%	83%	-1.10%

Among the countries least pressed to implement softswitches due to the need to replace aging equipment

Bangladesh	8.20%	60.10%	31.60%
Bulgaria	4.20%	39.30%	56.50%
China	7.10%	57.00%	35.90%
Egypt	13.90%	35.40%	50.70%
Portugal	48.90%	-9.70%	60.80%

Source: Dittberner Associates