Hybrid cloud: two competing models

This is a guest post by Gregory Ness, of CloudVelocity

Since the hybrid cloud has already been declared victorious, the next and perhaps even more relevant topic becomes a discussion of whose hybrid cloud will win.

Before you answer with the expected VMware, Azure, AWS, or Rackspace as a potential winner, think again. The winner may be none of those service providers.

Let me offer a counter-perspective: with hybrid cloud, innovators that can build connections between cloud services will win.

Before you dismiss this notion outright, let’s take a longer-term view. Since the days of the mainframe, technology companies have generated massive wealth by giving customers more and more control over data, where it’s delivered, where it’s stored, and even how it’s secured. Over time older technologies were ultimately cannibalized by newer technologies unleashing even greater potentials for management, control, access, delivery, and more.

Editor’s note: For a more in-depth discussion about how businesses are grappling with the “hybrid” cloud, come to VentureBeat’s CloudBeat conference, Sept. 9-Sept. 10 in San Francisco. We’ll be featuring executives from leading cloud vendors and customers — with eye to show how real businesses are adapting to the cloud. Register today!

VMware unleashed the latest IT revolution by decoupling hardware from apps and operating systems. Chief information officers were offered the ability to deploy more apps and operating systems on every server, helping them to use their gear more efficiently, which, commoditized the once mighty and influential server industry. VMware gave CIOs more control over their environments — enough to generate massive growth and profits over several years — and set the stage for the cloud revolution.

As the population of inexpensive infrastructure (from servers to sensors) explodes, the ability to provision and orchestrate quickly and broadly (versus the centralized and outdated mainframe model) becomes more economically powerful.  That takes us to the point I want to make about today’s cloud revolution.

The public cloud has become a great location for a minority of enterprise apps, most of them designed for the public cloud or residing on the cloud for software development and test purposes. The hybrid cloud has similarly become the enterprise-friendly vision of the cloud, because it allows the public cloud to be integrated seamlessly with the data center. Hybrid clouds can be very expensive to build, because of a vast gulf of manual processes typically required to extend or re-architect existing data center apps into the cloud.

Yet, the hybrid cloud is still only the beginning of the cloud revolution. We are about to see the emergence of two key and yet very different hybrid cloud strategies, one driven by “fences” and the other by “bridges.”

The hybrid cloud fence model
The early hybrid cloud successes will come from cloud service providers, plus Microsoft and VMware that build the best fenced offering. They will make it difficult for customers to change, through combinations of product, pricing and cloud migration process tradeoffs. They will have coercive controls over customers, giving them short-term competitive advantages that will spur growth, product development, and profit during some early crucial years. The existence of fences (especially extensive manual processes required to migrate apps between clouds) will give competitive advantage to the market leaders and the consultant and service economy that feeds them new apps/customers.

At some point a majority of the critical processes for an increasing population of apps (that once helped sustain the fence model) will be automated and apps will be able to easily migrate from data center to cloud, between clouds and back to the data center. Those who build bridges between the clouds will become more powerful, because they will monetize the traffic between the clouds. They will introduce dynamic new boundless operating potentials for IT.

The hybrid cloud bridge model

When the cost of migrating into and between clouds goes to less than 10 percent of the cost of operating an application in a single cloud or on premise, a tipping point will unleash a massive transformation of IT operating models. The fenced cloud will become obsolete, and enterprises will be able to leverage an abundant supply of infrastructure as a service (IaaS) options across regions, clouds, and international boundaries.

This is how I think the changes will take place, starting with the maturing of cloud migration tools into cloud migration and integration platforms. Early DevTest in the cloud tools (templates, nested hypervisors, image migration, and conversion tools) will be replaced by more powerful and comprehensive platforms that significantly reduce hybrid cloud deployment costs. That will then drive the use of the cloud for production apps, primarily for “pay as you go” disaster recovery. The use of the cloud for production apps, especially DR (or even augmenting DR with an extra layer of usage-based pricing protection) will drive massive cloud adoption for mid-market and second tier enterprise apps.

It will also fund the builders of the bridges, in the same way that the collision of enterprise apps with the Internet drove server load balancing into more profitable software-centric traffic optimization between servers and users. In the same way that IBM, Microsoft, Cisco, and VMware stood for more IT orchestration capabilities, the cloud migration and integration platforms that sit between clouds and data centers will unleash unprecedented IT increases in control and agility across a massive cloud infrastructure today being built by more than $50 billion in development spending.

Stated simply, the collision between the trillions spent in data centers and the tens of billions spent in clouds will be disruptive on a scale that few today will sufficiently appreciate. There will be a new generation of technology leaders who today might be easy to miss among the noise and mayhem being generated by the dinosaurs of tomorrow. Enterprises will enjoy unprecedented freedom of cloud, or unparalleled operating options. Those who try to trap them will lose.

Greg Ness is vice president of marketing at CloudVelocity, a provider of hybrid cloud software for the Global 2000. As a Future in Review (FiRe) conference panelist, he was among the first to point out network and security issues with virtualized IT infrastructures and then emerging cloud operating models, which led to the formation of the Infrastructure 2.0 Working Group with FiRe advisor Dan Lynch and a host of leaders in the networking industry. Greg’s Archimedius blog is one of the world’s most influential cloud computing blogs.