SDN, or software defined networking, is one of the hottest technologies in tech these days. But IBM is reported by the Wall Street Journal’s “re/code” to be shopping its SDN business. Selling its faltering server business to Lenovo is understandable, but why give up on SDN, a technology that has the potential to reshape data networks and is inspiring the next generation of connectivity in the view of many network experts. 

Is IBM (NYSE: IBM) management panicking after two years of declining revenue?

With earnings being boosted through big stock buybacks and cost cutting, has Chairman and CEO Ginny Rometty thrown in the towel on a network architecture that other firms such as Cisco and VMware are pursuing with vigor?

IBM wants out of the hardware business, as clearly shown by the $2.3 billion server sale announced last week to Lenovo.

Meanwhile, Big Blue is pouring more than $1 billion into “cloud” services and another $1 billion to “Watson” supercomputing with funds set aside for Watson-related startups.

So Big Blue wants to capitalize on the cloud. So does everyone else.

So Big Blue wants to turn Watson into a big dollar revenue producer rather than the answer to trivia questions about Jeopardy.

Big Blue also has set aside $1 billion to cover cost cuts and realignment. Nothing new about more layoffs and “resource actions” at IBM – unfortunately.

But SDN?

“Informally” Shopping for a Buyer

“IBM is said to have informally approached several companies in the networking business, including Dell, Hewlett-Packard, Cisco, Juniper and Fujitsu, to gauge interest. As IBM has recently spent big to expand its Watson cognitive computing unit as well as its SoftLayer cloud computing unit, the SDN unit is not seen as strategically important, according to a source familiar with the company’s thinking,” The WSJ says.

IBM wants $1 billion, sources told Re/Code.

But that price may be a problem – just as the dollars IBM wanted from Lenovo a year ago pushed back the x86 server deal until Lenovo got a price at its liking.

“One executive described IBM’s initial asking price of $1 billion as ‘pretty optimistic,’ Re/Code’s Arik Hesseldahl wrote.

Ben Kepes, writing at Forbes.com, was puzzled by the report.

“What we’re seeing is the emergence of a new strategy from IBM that sees it only try and compete where it has a compelling differentiator,” Kepes wrote. “The server business is one which has been decimated by cheaper OEM vendors over the last few years – IBM is a top shelf vendor with pricing to match – it can’t compete in the x86 server market so it was right to flick it on.

“But is that the case when it comes to SDN?”

(He also looks at IBM’s “cloud” strategy and what’s going on with Watson. It’s a story well worth reading.)

SDN Challenges vs. ROI

John Moore, senior director of innovation at MCNC in RTP, wrote this about SDN for WRALTechWire earlier this month:

“Software-Defined Networking (SDN) has captured the imagination of the networking community.

“At its heart, it’s both a simple and revolutionary idea – break apart the two major functions of the routers and switches that run the Internet and separate them onto different hardware platforms. The part that actually moves individual packets of information (the data plane) stays in the middle of the network and does its job, while the part that determines how those packets should be forwarded (the control plane) moves to a server that talks to the routers and switches (the controller).

“One advantage of this separation is that it opens interfaces between the component functions and allows for innovation by technology companies and the open source software community. The controller component in particular is ripe for developing compelling new products and building market share.”

However, Moore went on to note that SDN still faces technology development hurdles.

Perhaps IBM decided not to invest further in a technology that isn’t guaranteed to provide long-term payback – a return on investment.

Giving the pressing need to cut losses and drive more revenue immediately, perhaps IBM is making a wise choice if SDN is sold.

But for the long term, Big Blue could be making a big, costly mistake, especially as demand for faster, better networks – land-line and wireless – as well as bigger, faster data centers continues to grow.