In their efforts to establish context, many of the journalists and vendors we speak with begin by comparing the cloud to ancestral technologies like grid or utility compute offerings. Many, if not most, believe that is representative of the possible upside for the cloud and cloud based offerings.
Certainly those of this mind set may yet be proven correct, irrational exuberance being a historically common failing in this industry. But for my money, such analogies undersell – dramatically – the cloud’s horizontal appeal.
Shortly after Network.com launched in 2006, I had the opportunity to speak with a few of the folks from Sun about the offering. Their first question to me was both expected and straightforward: “Have you signed up for Network.com yet?” My answer was just as simple: “No. Why would I?”
And indeed, why would I? With all due respect to Sun’s ability get to market early, Network.com had little to offer an individual technologist. Because it was focused on batch operations that could be easily outsourced rather than general workloads, its applicability to my needs was questionable. This belief was further validated in the specialized nature of the customers of the service I spoke with: oil and gas exploration, fluid hydrodynamics research, digital animation rendering and so on. The typical customer case study from Network.com was as impressive as it was exceptional. Sun’s Grid was an early, impressive achievement, but one targeted to as narrow an audience, in my view, as its E10K machines – but without the similar margins.
Sun was hardly alone in its lack of delivery here, of course. If anything, the soon-to-be-Oracle-acquisition was ahead of the curve: when a major platform services market is pioneered by an online retailer, it’s safe to say that the vendors missed a trick. Microsoft’s Ray Ozzie admitted as much to Om Malik, saying in March of last year that the market “really isn’t being taken seriously right now by anybody except Amazon. They’ve done the world a service by putting out there some fairly provocative, interesting services.”
Why are their services interesting? Their appeal. Unlike the grid and utility efforts that preceded them, the cloud was more horizontally applicable, and – if anything – is becoming more so every day. Grid computing is demonstrably useful to the Exxons and Pixars of the world while cloud services like storage are potentially relevant to everyone from a Fortune 50 to your neighbor backing up his music collection. That alone differentiates cloud from other important technical trends such as SOA. Like operating systems, the cloud can play in consumer and enterprise with near equal facility, making it a rare beast indeed.
The implications for provider and vendors are, of course, varied, but if you combine the appeal with the low barriers to entry, what you have is what I’ve been telling customers: an opportunity not to fight house to house. Most enterprise markets are supersaturated, with customer acquisition and retention a costly chore thanks to the overhead of the requisite sales, marketing and support staffs. The cloud alone is no panacea for that condition, of course, for at certain purchasing levels it will be forced to jump through the same hoops as every other enterprise technology offering.
Like open source, however, the cloud can – by virtue of its applicability and accessibility – flow into the enterprise basement like flood waters. Rather than fight it out house to house, sales call by sales call, the opportunity exists to have an offering that sells itself. And better, from a vendor perspective, may be difficult to decommit from, if only because of the pains of migrating significant volumes of data.
But to get there, you have to do what Amazon did: take the market seriously.
Disclosure: Microsoft and Sun are RedMonk customers, while Amazon is not.