Why Siemens Sold Unisphere

When Siemens sold its Unisphere subsidiary to Juniper Networks Inc. (Nasdaq: JNPR) earlier this month, a lot of attention was focused on whether the deal was a good one for Juniper (see Juniper Nabs Unisphere for $740M).

This week, the other side of the story emerged -- why Siemens AG (NYSE: SI; Frankfurt: SIE) sold Unisphere Networks Inc. and how it fits in with a refocusing of strategy at the German giant (see Siemens Outlines Next Gen Strategy).

In a nutshell, it comes down to this: Like most telecom equipment manufacturers, Siemens Information and Communications Networks Inc. is having to cut back, so it's putting a lot of its chips on becoming the World No. 1 in softswitches.

Why softswitches? Or should that be: What the heck are softswitches? Let me explain.

The market for "big hummer" traditional telephone exchanges is finally undergoing the same sort of revolution that killed off all but one mainframe computer manufacturer 20 years ago, following the arrival of the PC.

In essence, the call control in old-fashioned telephone exchanges can now be hived off into some software running on a server, and the switching matrix can be replaced by IP networks. There's no need for honking great bits of hardware any more. All that's needed are routers and a bunch of gateways for linking together old and new stuff.

On the software side, application program interfaces that open things up for independent software developers are being standardized. This is likely to reduce prices and stimulate a lot of innovation on the applications front.

These developments have been going on for several years already and haven't really got past the stage of talking and testing. But Siemens now thinks the time has come for some serious, large-scale deployments. A growing number of old telephone exchanges are reaching the ends of their lives, according to Christof Wahl, president of wireline network communications in Siemens' ICN group. As a result, he contends, carriers will take the opportunity of shifting everything onto Internet Protocol (IP) infrastructure to cut their operating costs and pave the way for providing a much greater variety of services.

Wahl cites some examples of carriers that are already taking the plunge with the everything-over-IP approach, using Siemens' Surpass family of softswitching products. These include China Telecom, Finland's Sonera Corp., and other operators in Greece, Denmark, and Taiwan.

So, how does the sale of Unisphere fit into Siemens' goal of being the biggest player in the softswitch market?

First off, it's important to note that Unisphere had two product lines -- edge routers and softswitches -- up until early May, when Siemens took over the softswitching products, leaving the edge router business to be sold to Juniper for stock and cash worth $740 million (see Siemens Absorbs Unisphere's Voice Biz).

As a result of this, Siemens will own a 9.7 percent stake in Juniper once the transaction is completed, and it will also become a big reseller of Juniper core and edge routers (some of them being ex-Unisphere). In other words, Siemens and Juniper have got a close relationship -- one that Siemens intends to exploit in order to strengthen its offerings to carriers wanting to run their voice services over IP infrastructure.

Siemens' pitch on this goes as follows: It already makes the industry's most reliable big hummer telephone switches, and achieving this in networks serving millions of people requires specialized knowledge garnered from decades of experience.

Siemens says that this gives it an edge in delivering equivalent reliability in a softswitch environment, and that the deal with Juniper now means it can ensure that this extends into the underlying IP infrastructure. Specifically, lots of testing and development work is being conducted on combinations of Siemens' softswitch products and Juniper's core and edge routers to ensure that they're just as reliable as traditional telephone networks, according to Wahl.

The only way of proving Siemens' reliability-with-scaleability claim is to build an actual network serving millions of users -- and nobody's going to do that just to see whether it works. In other words, this could be mainly market-speak, aimed at spreading FUD (fear, uncertainty, and doubt) among carriers thinking of using voice-over-IP gear from vendors with less experience at building really big networks -- such as Cisco Systems Inc. (Nasdaq: CSCO).

Another reason Siemens sold off Unisphere might be that it didn't have much faith in its long-term future. Siemens is adding capabilities for handling voice-over-IP to access equipment such as DSLAMs (digital subscriber line access multiplexers). It's also adding modules to its Sonet (Synchronous Optical NETwork) and SDH (Synchronous Digital Hierarchy) gear so that it can carry data more efficiently. In the long run, this might reduce requirements for routers.

Three other little bombshells were dropped quietly at Siemens' international press briefing this week:

  • Like Marconi PLC (Nasdaq/London: MONI), Siemens has put its passive optical networking (PON) developments on the back burner (see Salira Prompts PON Questions).

  • Siemens sees Multiprotocol Label Switching (MPLS) as a protocol for enterprise networks. Wahl says it simply won't scale to the number of connections required in a really big public network. Siemens is working with several German universities on a project called "King" (from the German for Components for Next-Generation Internet) to develop an alternative connectionless protocol.

  • Siemens is negotiating a reseller agreement with an unspecified manufacturer of a sub-wavelength grooming switch. As this is clearly not a direct competitor such as Alcatel SA (NYSE: ALA; Paris: CGEP:PA), Lucent Technologies Inc. (NYSE: LU), or Nortel Networks Corp. (NYSE/Toronto: NT), it could be Ciena Corp. (Nasdaq: CIEN).

    — Peter Heywood, Founding Editor, Light Reading
  • <<   <   Page 2 / 2
    photon_tim 12/4/2012 | 10:18:55 PM
    re: Why Siemens Sold Unisphere Hi Kangaroo,

    Yes, I agree changing strategy is better than becoming extinct. However my point was that Siemens is giving up the "boxes" piece of the brave new world. (i.e. the router/media gateway/IP hardware). In my view a softswitch is 95% software and some bits of standard hardware (I may be wrong?).

    In the IBM analogy it seems to me that Siemens decided that Mainframes are about to die (right so far), but instead of building servers and PCs (i.e. HW) they decided to compete with software only (OS/2 vs Microsoft?).

    photonicGuru 12/4/2012 | 10:18:53 PM
    re: Why Siemens Sold Unisphere Why seimens sold unisphere?....we already discusssed that in detail a week before and some of the posting are very much in line with our discussion...revenue streams and finding new markets.

    As far as seimens startegy goes, their conclusion about sub wavelength grooming and MPLS non scalability for large carrier class network are good. But, i do differ with the starategy of putting PON network on backburner....if one consider the sparse density of future broadband users in US, it may be okay for atleast 2005 .But in terms of Europe and some big asian cities, the density overkill in metropolis, reducing cost of passive optical components and new packaging technologies in pipeline making PON systems more attractive for large scale deployment....the carriers and vendors still haven't lost their interest in this even after 10 years becoz' some of them know very clearly that their is no way out in the end with consistent overload of information flow at the user's end (access network). Same economics also driving the softswitch technology, the price of vanilla switch per port hangin' below $50 (even less if you want to buy from auction houses....excess inventory of Nortel!!). There is no profit left for vendors in this line of business. Then, the option left with marketing gigs and techie gigs to promote all IP architecture and selling new gears to their customers ...i m not against softswitch...pretty neat technology and signalling architecture...ready to roll for new deployments but they need to cut their technology price/performance line quickly for large scale deployment!!

    Gandalf 12/4/2012 | 10:18:48 PM
    re: Why Siemens Sold Unisphere photon_tim wrote:

    While softswitches are a good idea for a start-up, I do not buy that this is a smart strategy for a biggie like Siemens (Lucent, Nortel).

    If I may disagree on this point - each one of these companies you are listing has at least one (if not multiple) program(s) in place to create their own softswitch. On average I would estimate they each have anywhere from 300 to 600 engineers working hard to catch up with those "start-ups" that have taken a substantial lead (Sonus, Santera, Telica). That's a burn rate of some $35Mil to $90Mil per year and somehow I cannot see how could they justify that without a solid business plan - meaning they believe in softswitch technology. Every major ILEC issued at least one RFI/RFP on softswitch technology in the past 12 months and they all have one or more softswitch vendors in their labs testing.

    The current price per port is typically around $50, but I would agree it will come down under competition pressure. I don't think it will hit $20 any time soon, though.
    Gandalf 12/4/2012 | 10:18:45 PM
    re: Why Siemens Sold Unisphere Addendum - I really should have added a quote from today's Supercomm event:

    Carriers and enterprises appear to concur. A February 2002 study released by Cahners InStat/MDR projects that worldwide softswitch sales will increase from last year's $125 million to $1.32 billion in 2006, a compound annual growth rate of slightly more than 60 percent."
    <<   <   Page 2 / 2
    Sign In