Should Cisco Buy NSN?

5:55 PM -- What if Cisco Systems Inc. (Nasdaq: CSCO) decided to acquire the optical division of Nokia Networks ?

The question came up a couple of weeks ago in the optical networking group on LinkedIn. It was just a what-if scenario posed by Gabriel Kerner, a business development manager at Huawei Technologies Co. Ltd. .

But Wednesday morning, Bloomberg reported that NSN is indeed "negotiating with buyers to sell some assets," as CEO Rajeev Suri is quoted saying. That could include optical networking (or Carrier Ethernet -- see Can NSN Offload Its Carrier Ethernet Assets?).

So, it's worth asking: Should Cisco be tempted?

Consider what Cisco has to gain. Kerner's original point is that service providers seem to like the idea of a combined management system -- one set of software to control the optical and routed networks. This is what motivated Juniper Networks Inc. (NYSE: JNPR) to partner up with NSN and, separately, with ADVA Optical Networking . (See Juniper's Packet-Optical Spells M-P-L-S and Juniper Finds Another Optical Partner.)

Cisco is clearly revving up optical again. The company is clearly proud of its CoreOptics acquisition and its 100Gbit/s promise. Then there's Lightwire Inc. , the silicon photonics startup Cisco is acquiring. (See Lightwire Points Cisco Toward 100G and Cisco Renews Optical Focus With CoreOptics.)

Even better, from Cisco's point of view, is that NSN could be bought with offshore cash. It would be one more chance for Cisco to needle the U.S. government about the tax on repatriation. Cisco is already saying its hiring and acquisition patterns will gravitate to international sites because of that issue. (See Chambers Floats His Stimulus Plan.)

Cool, huh? What's not to like?


  • The deal would probably come with thousands more employees than Cisco wants. And laying off workers in Europe is like prying beer from the hands of football fans -- even when it's doable, it's messy.

  • NSN overlaps Cisco substantially. Even as an assets-only purchase (assuming NSN would go for that), it's likely the biggest motivation would be defensive.

  • Cisco doesn't like big acquisitions in the first place. Scientific Atlanta comes to mind as a comparison, but that's a business Cisco was entirely absent from. For the amount of integration work it would take, NSN doesn't seem like an ideal target.

    I have no idea whether Cisco is even interested in NSN. I'm just amused by the way the pieces do and don't fit together. What would your pros and cons be?

    — Craig Matsumoto, Managing Editor, Light Reading

  • ninjaturtle 12/5/2012 | 5:40:59 PM
    re: Should Cisco Buy NSN?

    INFN would be a huge add to CSCO's optical business. Don't know if INFN is interested in being acquired. Any thoughts?

    spc_markl 12/5/2012 | 5:40:58 PM
    re: Should Cisco Buy NSN?

    A Cisco-Ciena deal would make the most sense in that the former would acquire a huge installed base of optical switches.  Hopefully, Ciena's CEO is no longer in a powerful position to stop such a transaction.

    Mark Lutkowitz

    paolo.franzoi 12/5/2012 | 5:40:57 PM
    re: Should Cisco Buy NSN?


    I agree with you Mark that Cisco-Ciena makes sense.  Would a Cisco/Tellabs deal for their optical group makes sense?  Been thinking the best thing for Tellabs is to break it up and sell off the pieces.  E might be interested in the wireless backhaul stuff and the rest can (after that and optical) could go to PE to drive cash out until it dies completely.


    spc_markl 12/5/2012 | 5:40:56 PM
    re: Should Cisco Buy NSN?


    For better or worse, I just do not see Tellabs selling out anytime soon.  At the end of the day, it is a family-run business and they will likely hang in there as long as possible.


    Flook 12/5/2012 | 5:40:53 PM
    re: Should Cisco Buy NSN?

    Ditto. Would be better for Cisco to go with a US-based company because across the pond acquisitions are bound to be problematic due to differences in culture, in this case the distinct culture of the two vendors, not to mention culture from the broader national perspective.

    Pete Baldwin 12/5/2012 | 5:40:47 PM
    re: Should Cisco Buy NSN?

    I wonder if it's more likely Cisco would plomb the optical startups -- a few do exist -- before going after an established company.

    It would be cheaper and smaller, more to Cisco's liking. And it would net a cutting-edge technology set and a high density of engineers.

    Infinera would get them some good technology but also a fab. I don't think Cisco could deal with that.

    Ciena seems too big, and I wonder how their whole packet-optical strategy would mesh (or not) with Cisco.  Tellabs...hm. A piece of Tellabs might be interesting, but would it have the level of innovation Cisco wants? (And as Mark says, it might not be doable.)

    I like Flook's point about a US acquisition in general, but Cisco's dead-set on spending that overseas cash on overseas properties.

    msinisca 12/5/2012 | 5:40:46 PM
    re: Should Cisco Buy NSN?

    To Gabriel Kerner's original point:  "that service providers seem to like the idea of a combined management system -- one set of software to control the optical and routed networks", yes everybody wants this but when has this ever happened?

    Cisco bought Cerent (15454 line) and Pirelli (15800) in 1999 and later transferred technology (cannibalized) to the 15454 product line. To this day the Cisco router (IOS, NX-OS), switch (Catalyst) and optical (CPO, CTM, ONS) platforms still have separate EMS and NMS.

    I am not singling out Cisco. ADVA did the same with the Movaz acquisition in 2006. Lucent acquired Ignitus, Chromatis and about 10 other networking vendors from 1997 to 2001, all were stripped of technology, down sized and introduced into existing product lines. Zhone has done this with a dozen companies. In the test equipment and component space JDSU is the serial acquirer bar none.

    The reality is that corporations acquire companies to add technology and IP sans R&D time and cost, gain market share (eliminate competition) or enter into new markets. All of this is driven by profit. Full integration (HW and SW) of multiple product lines are usually cost prohibitive and more often technologically infeasible (differing form factors, SW schema, backward support). Continued development across several product lines is too expensive and eventually one line is scrapped (the acquired) and migrated. In my experience, most product integrations involve little more than a new powder coat in the acquiring company’s trademark color and shiny new stickers.

    Integration of the hit7300 and 15454 will happen only at a rudimentary NMS monitoring level and eventually the hit7300 will be phased out (remember the 15540 ESP). I believe the biggest motivation for Cisco, as was the case with ADVA, is incorporating NSN's GMPLS control plane and packet switch fabric into the 15454 and secondarily gaining European market share.

    Sign In