As the months wear on and confidence in public clouds begins to slowly creep skyward there is no doubt that there will be a public cloud provider boom. The questions are, of course, just how big will it be and moreover, how long until a major industry shakeout occurs? After all, in the tech industry anyway, every boom appears to be accompanied by an equally fierce bust.
Last week yet another public cloud provider emerged from the ether, this time in the Netherlands, which is likely a much more hospitable environment since there is a weaker tendency to rely on international providers of anything, let alone coveted enterprise-specific data. This company, called Oxilion,announced its virtualization partnership with Red Hat and in doing so gave some insight into how new cloud providers are making critical decisions that will affect not only their ability to get off the ground, but what they will be able to offer potential customers down the road.
In the U.S., all of this seems like a rather nervy thing to do—competing with public cloud giants who are some of the first thus most established in the space, not to mention who are also usually household names. According to Scott Crenshaw, Vice President and General Manager of the Cloud Business Unit at Red Hat, this movement toward public cloud provider growth (or sprawl) is on the horizon. When asked if he saw a large influx of startups with toes to public cloud line, Crenshaw responded:
We see many companies planning to enter the market to become cloud service providers. They include telecom service providers, IT service companies, and hosting companies. And the interest spans the globe, in every continent, with one exception: I haven’t seen any from Antarctica yet. There are a common set of requirements for their infrastructure: the ability to provide Quality of Service (QoS) to deliver customers SLAs, scalability, high performance (which reduces hardware cost), and security in multi-tenant environments. This is exciting new ground for the industry.
Crenshaw added that we will be seeing “a proliferation of public clouds over the next few years” and accordingly, at least from Red Hat’s business perspective, “users will demand better service levels, and consistency between the stacks they use in their data center, and those available on the public cloud. In the end, cloud providers will be driven by scale economies. Those that don’t have scale, or don’t carve out a clearly differentiated niche, will struggle. But I think that’s several years away. In the meantime, we’ll see fast expansion. We’re in the very early days of this industry.”
As with everything the mighty concept of economies of scale is at the root of the discussion. But the question really is, “how many cloud providers do we need” and moreoever, what kind of boom are we going to see so we can get an advance feel for the bust. With everyone marketing their services as “cloud” related in one way or another, when will people tire of the term and when they do, will they revert to the steady EC2, Google, or others?
Cost Drives Cloud Adption, Cloud Boom, and Cloud Burst
When I refer to a cloud burst in this context, it should be clear I am not talking about hybrid cloud models. Think bubbles. Think the 1990s. Think bust.
Since cost is certainly one of the reasons that such a proliferation can exist. I asked Crenshaw the multi-layered question: To what degree is initial start-up cost the deciding factor for firms like Oxilion who seek you out? Is there where your advantage lies and if so, how does the pricing work out over the long term (so clarify how while startup costs might be negated, over the long term the expense continues to mount, making that initial savings no longer a valid point)?
According to Red Hat’s VP and Manager of its cloud business initiative, “Cost is certainly important for cloud providers. Software cost is just one component. Hardware cost is another. By providing more efficient virtualization, with Quality-of-Service capabilities built-in, cloud providers can buy less hardware when they use Red Hat Enterprise Virtualization, spending less on hardware acquisition and administration.
Crenshaw continued, “As important is the scalability of the infrastructure. Clouds are elastic — they have to scale up and down. And of course, security is another critical factor; only Red Hat has policy-based security built in to the core of the kernel.”
On the Inevitable Thinning of the Herd
Rooted in Crenshaw’s statement is the concept that unless cloud providers work diligently to attach to a niche they will struggle. They must offer customers a level of service that is capable of exceeding the big names in the industry—a tall order for a small company, indeed. As the year goes on it will be interesting to see how many new providers we can add to a list that for now (and perhaps always) will be dominated by the giants.
I wonder if cloud adoption where it matters–in the enterprise and up–will keep up with clamor of thousands of startups and established companies who want your attention. We live in some very interesting times, indeed.