March 19, 2014
March Madness – Five Years Ago
It was five years ago, March 2009, when Cisco formally announced “Project California,” its (possibly intentionally) worst-kept secret, as Cisco Unified Computing System. At the time, I was working at Hewlett Packard, and our collective feelings as we realized that Cisco really did intend to challenge us in the server market were a mixed bag. Some of us were amused at their presumption, others were concerned that there might be something there, since we had odd bits and pieces of intelligence about the former Nuova, the Cisco spin-out/spin-in that developed UCS. Most of us were convinced that they would have trouble running a server business at margins we knew would be substantially lower than their margins in their core switch business. Sitting on top of our shiny, still relatively new HP c-Class BladeSystem, which had overtaken IBM’s BladeCenter as the leading blade product, we were collectively unconcerned, as well as puzzled about Cisco’s decision to upset a nice stable arrangement where IBM, HP and Dell sold possibly a Billion dollars’ worth of Cisco gear between them.
Five years later, HP is still number one in blade server units and revenue, but Cisco appears to be now number two in blades, and closing in on number three world-wide in server sales as well. The numbers are impressive:
- 32,000 net new customers in five years, with 14,000 repeat customers
- Claimed $2 Billion+ annual run-rate
- Order growth rate claimed in “mid-30s” range, probably about three times the growth rate of any competing product line.
Better technology can still be disruptive. The perception that a technology segment is even remotely stable can be dangerous to incumbents – in retrospect, going into the new 21st century a certain collective complacency among the market leaders about the implied oligopoly of suppliers and the neatly assigned silos of storage, networking and servers. But as the two leaders, IBM and HP, began to build out blade servers, each of which effectively removed several dozen potential switch ports from Cisco’s TAM, Cisco began to eye the immediate market adjacency of servers, and the die was cast.
The net result was that Cisco built a better mousetrap and the world beat a path to their door. UCS was a brilliantly conceived product that took the innovative concept of virtualizing much of the physical server identity and I/O mappings that made HP’s BladeSystem so successful and extended them even further, along with (no surprise here) deep integration with Cisco networks. The UCS Service Profile, the software definition of the server configuration, identity and even some aspects of run-time behavior was a more powerful abstraction than others offered, and combined with a truly excellent integrated management console, offered real productivity gains compared to competitors.
Additionally, Cisco was more aggressive going to market than any of its competitors anticipated; probably subsidizing server margins in selected deals with margins from other product lines, and investing in a quality overlay sales force and a channel program which now approaches 2000 partners.
The New Reality
The world of x86 servers now has four significant global players rather than three – Cisco is not going to exit the server business, and UCS will remain an integral part of their portfolio. Future developments are hard to predict, but my bets are that Cisco will stick close to their mainstream enterprise roots and that while they may build servers that are closer to IBM and HP “density-optimized” servers for big data and analytics, they will not attempt to dilute their offering with lower-margin hyperscale or micro-server offerings. They will continue to expand their cloud services and data center business as well, and have put a marker down in high-performance storage with their Whiptail acquisition.
Key Takeaway – Cisco has taken significant share from the major competitors, and with the growth rates they are exhibiting they will probably continue to do so until such time as their growth rate begins to approximate the industry average.[i]
Are we done? Could we see another disruption of established relationships in the core infrastructure market or is it now in a deep equilibrium that is guaranteed for another couple of decades? While Cisco, HP and IBM dance their Gavotte in the server market, customers continued to enthusiastically buy tons of storage. Is it possible that a large storage vendor, one who has survived every attempt by server vendors to usurp its storage franchise and already owns leading virtualization technology and has begun to package servers with its storage for selected pre-integrated offerings, might decide that the market for servers would be a nice adjacent market? The basic question they would have to answer would be whether the offended former partners have a viable alternative storage partner that they could divert customer dollars to as a competitive response? My guess is no – all experience tells us that enterprise customers are in general more loyal to their storage than to their servers.
So my nomination for a possible disruption would be for EMC to enter the server business, probably not initially as a standalone server vendor, but by aggressively selling bundled server and storage offerings into the most rapidly growing segments of the market, primarily big data and customer analytics. They have already built the machinery with their own bundled VMware converged infrastructure offerings, have the software IP to make very functional bundles, and could quietly accelerate the pace of introduction of such bundles over the next few years, and it will be easier for them to add servers to their offerings than it will be for former partners to find acceptable substitutes.[ii]
[i] Every time I say that Cisco has taken market share from “Company A”, “Company A” comes back to me with a vigorous statement that “our revenues have grown”, and occasionally some creative massaging of market numbers that indicates that they now sell more servers every Tuesday in Patagonia than any other competitor. But the aggregate numbers don’t lie. Cisco has been growing significantly faster than the overall server market, which has been in kind of global funk for a while, and if Cisco was not around, then “Vendor A” would have grown much more. It’s hard to prove, but my guess from talking to Forrester clients is that about two thirds of Cisco’s growth came at the expense of HP and the rest mostly from IBM’s potential revenues. Another response is “we don’t see ourselves losing to Cisco in competitive situations”. Maybe so (and maybe not), but not all situations are competitive. I have personally talked with customers of “Vendor A” and “Vendor B” who simply went out and bought big gobs of UCS without ever making it a formal competition – they had years of experience with the incumbent, evaluated UCS and purchased them, in some cases without the incumbent vendor’s sales team even realizing their account was in jeopardy. Maybe that account team didn’t subsequently advertise to their management that they had lost an account since it wasn’t a “competitive” loss. If it was an account serviced through channels it would have been effectively invisible to the vendor’s management team.
[ii] MAJOR DISCLAIMER – This is my speculation, pure and simple. I have absolutely no real data to back this scenario up. In fact, if it was going to happen I would probably be the last to know.
- cloud computing
- converged infrastructure
- data centers & data center networking
- infrastructure & operations