Like Om Malik, I received and read the memo that Cisco distributed internally regarding the company’s plans for spin-in Insieme and software-defined networking (SDN). Om has published the memo in its entirety, so there’s no need for me to do the same here.
As for Insieme, the memo informs us that Cisco has made an investment of $100 million in the “early-stage company focused on research and development in the datacenter market. It also notes that Insieme was founded by Mario Mazzola, Luca Cafiero, and Prem Jain in February 2012, and that “Cisco has the right to purchase the remaining interests of Insieme, with a potential payout range of up to $750 million that will be based primarily on the sales and profitability of Insieme products through Cisco.”
Cisco emphasizes that Insieme’s product-development efforts are “complementary” to its current and planned internal efforts, and it notes that further details regarding Insieme will be disclosed in “Cisco’s upcoming 1oQ filing in May.”
Mystery No More
But we don’t have to wait until then to discern how Cisco will position itself in relation to SDN and programmable networks. If we were in need of additional clues as to how Cisco will play its hand, the memo contains more than enough information from which to deduce the company’s strategy.
As far as Cisco is concerned, there isn’t actually anything new to see in SDN. This is where the marketing battle over words and meanings will ensue, because Cisco’s definition of SDN will bear an uncanny resemblance to what it already does today.
In the memo, Padmasree Warrior, Cisco CTO and co-leader of engineering, makes the following statement: “Cisco believes SDN is part of our vision of the intelligent network that is more open, programmable, and application aware—a vision in which the network is transformed into a more effective business enabler.”
It’s an ambiguous and innocuous opening salvo, and it could mean almost anything. As the memo proceeds, however, Cisco increasingly qualifies what it means by the term SDN. It also tells us how Insieme fits into the picture.
Here’s what I see as the memo’s money shot:
“Because SDN is still in its embryonic stage, a consensus has yet to be reached on its exact definition. Some equate SDN with OpenFlow or decoupling of control and data planes. Cisco’s view transcends this definition.”
If you want the gist of the memo in a nutshell, it’s all there. Cisco will (and does) contend that the “decoupling of control and data planes” — in other words, server-based software deciding how packets should be routed across networks — does not define SDN.
This should not come as a surprise. It’s in Cisco’s interest — and, the company will argue, its customers’ interests as well — for it to resist the decoupling of the control and data planes. You won’t get ridiculous hyperbole from me, so I won’t say that such a decoupling represents an existential threat to Cisco. That would be exaggeration for effect, and I don’t play that game. So let me put it another way: It is a business problem that Cisco would rather not have to address.
Could Cisco deal with that problem? Probably, given the resources at its disposal. But it would be a hassle and a headache, and it would require Cisco to change into something different from what it is today. If you’re Cisco, not having to deal with the problem seems a better option.
Later in the Cisco memo, the company tips its hand further. Quoting directly:
While SDN concepts like network virtualization may sound new, Cisco has played a leadership role in this market for many years leveraging its build, buy, partner strategy. For example, Cisco’s Nexus 1000V series switches—which provide sophisticated NX-OS networking capabilities in virtualized environment down to the virtual machine level—are built upon a controller/agent architecture, a fundamental building block of SDN solutions. With more than 5,000 customers today, Cisco has been shipping this technology for a long time.
“SDN plays into at least two of Cisco’s top five priorities—core routing/switching and data center/virtualization/cloud,” says Warrior.
Cisco has the opportunity to shape and define the SDN market because it is still perceived as an emerging technology, Warrior says. In fact, Cisco innovation will be much deeper than just SDN.
Cisco is operating from established positions of strength, which include the scale of its operating systems, superior ASICS, unique embedded intelligence, experienced engineering expertise, and an expansive installed base—most of which has no interest in completely replacing what it has already invested in so heavily. “
Pouring the Grappa
So, Cisco’s future SDN, including whatever Insieme eventually delivers to market, will look a lot like the “SDN” that Cisco delivers today in the Nexus 1000V series switches and the NX-OS. When one considers that some engineers now on the Insieme team worked on the Nexus 1000V, and that Insieme is licensed to use the NX-OS, it does not take a particularly athletic leap of logic to conclude that Insieme will be building a Nexus-like switch, though perhaps one on steroids.
Insieme, as I’ve written before, will represent an evolution for Cisco, not a revolution. It will be fortified switching wine in an SDN bottle. (Mario Mazzola is fond of giving Italian names to his spin-in companies. He should have called this one “Grappa.”)
Commenting on Cisco’s SDN memo and the company’s decision to tap spin-in venture Insieme as a vehicle in the space, Om Malik interpreted it as “a tactical admission that it (Cisco) has become so big, so bureaucratic and so broken that it cannot count on internal teams to build any ground breaking products.”
Bigger This Time
That might be an accurate assessment, but it’s also possible to see Insieme as business as usual at Cisco. Clearly Cisco never retired its spin-in move, as I once thought it did, but merely put it into prolonged sabbatical, holding it in reserve for when it would be needed again. Malik himself notes that Cisco has gone to the spin-well before, with this particular trio of all-star engineers now involved in their third such venture.
For good or ill, maybe this is how Cisco gets difficult things done in its dotage. It calls together a bunch of proven quantities and old engineering hands and has them build a bigger datacenter switch than they built the last time.
Is that SDN? It’s Cisco’s SDN. The company’s customers ultimately will decide whether it’s theirs, too.