By all appearances, 2012 seemed to be a good year for cloud computing. In addition to being the darling of the media for all of last year, the cloud gained a lot of ground in terms of the number of businesses that have adopted cloud computing in some form.
But despite this traction, the cloud is still largely a spectator sport for providers from a financial perspective. Even by optimistic estimates, the market has realized less than 1% of its total opportunity.
Cloud providers or prospective providers are concerned; how can a service already be turning into a commodity when it's hardly begun to succeed? Fortunately, key improvements on both the buyer and seller sides of a cloud transaction will drive the future of cloud computing in 2013.
Five changes stand out as major trends cloud providers should be following this year.
1. PaaS becomes king
First among the cloud drivers is a shift toward Platform as a Service (PaaS) as the preferred cloud services model. Infrastructure as a Service (IaaS) products dominate the market today, but they're already falling prey to crushing price and margin pressures; moreover, it's clear that the IaaS business proposition isn't easily articulated anymore. But more significant is the fact that both cloud-literate buyers and cloud providers are asking, "If the cloud is so different from other service delivery models, why won't it spawn its own flood of cloud-specific applications?"
If applications are at the top of the cloud stack and servers are at the bottom, then IaaS is a minimalist entry into cloud computing. Costs accumulate for buyers as they go up the stack, but their potential savings follow the same trajectory. That is, a cloud service can only displace the cost of maintaining the same resource on-premises -- meaning that IaaS only displaces hardware and facilities costs for buyers.
A PaaS cloud with custom-built platform services and APIs can target new applications that are cloud-dependent. This will be the most transformational thing to ever hit the cloud.
Therefore, Software as a Service (SaaS) displaces the most costs and IaaS the least. But SaaS requires an extensive application software repertoire, and so cloud providers are turning to developers and integrators to create those higher layers. As they do so, providers realize that by offering more than just virtual machine (VM) hosting, they can add value to these partnerships and create a more uniform platform operations framework, thereby enabling providers to reduce their own cloud support costs.
The cloud is already breaking away from its heritage as a primary mechanism for cost-and-margins management. Recently, PaaS has enabled the cloud to begin evolving into something that could be called a "virtual cloud operating system" -- and many of them are just that. After all, every time a provider offers a new cloud-based service, it has effectively created a virtual cloud operating system for it. Today, most PaaS services mirror a conventional operating system (OS), but in the future, different providers' PaaS environments will diverge from each other and from today's conventional OSes to create truly cloud-based OSes.
Amazon Web Services, the leader in IaaS, has been enhancing its services beyond IaaS, most recently with the launch of its Redshift data warehousing cloud service. Microsoft has added Azure application programming interfaces (APIs) to Windows Server, making it possible for third parties to host Azure-compatible clouds. Red Hat has introduced a PaaS product, OpenShift, and Amazon has a Java VM-hosting platform optimized for the cloud. All of these advances ease cloud migration and encourage developers, and that's critical for fostering a rich set of services and benefits.
Developments in this arena will move further -- and faster -- in 2013. Telefónica Digital, for example, plans to build on its popular TU Me unified communications service by making the core APIs, called TU Core, available to developers.
CIMI Corp. estimates that the revenue opportunity for simply migrating legacy applications to the cloud is a mere one-tenth of the total cloud opportunity. But by creating their own custom platform services and APIs, operators are creating a PaaS cloud that can target more than just the hosting of current applications; it can also target completely new ones that are cloud-dependent. This approach will be the most transformational thing to ever hit the cloud.
2. Buyers get smarter about cloud
The second factor driving cloud change in 2013 is reaching critical mass in cloud-buyer literacy. When less than a third of buyers in any market are able to confidently understand and articulate the value of a product or service -- never mind plan a deployment project -- the sales process is frustratingly difficult, project failures are high and growth is slow. In 2012, the buyer literacy rate for cloud services was less than 25%, but our modeling shows it will reach that critical one-third mark by the end of 2013.
The smarter buyer community means that cloud services will for the first time be driven by focused market pressures and specific market needs. All of this will multiply opportunities for aggressive cloud operators, pressure those with substandard offerings and drive a wave of cloud consolidation.
3. SOA cannot be ignored
Third on the list is growing recognition of the role of software-oriented architecture (SOA) in native-cloud services. One of the glaring omissions in long-term cloud strategy has been a model for how applications would develop to exploit the cloud -- not just run on it. More software vendors are now positioning SOA, RESTful (that is, Web-like) APIs and orchestration as a means of not only creating native cloud applications, but also for facilitating the hybridization of public cloud services with internal IT.
4. The network evolves
Driver number four is the improved integration of network services into the cloud. Software-defined networking and network virtualization are often intermingled, but they're in fact two separate initiatives that provide for both cloud control of network connectivity and performance. OpenStack's Quantum API offers direct control for virtual networks, and a growing number of DevOps tools are helping "operationalize" application lifecycle management and deployment in public, private and hybrid clouds.
5. Don't let Amazon scare you
The final driver for 2013 is competition and the need for differentiation. No provider wants to enter a market that's continually facing shrinking prices and margins, and yet even a major Tier 1 operator said in 2012 that the cloud would soon replace virtual private networks as the most common service for business customers. To create the "new cloud" and make it profitable, providers are exploring and expanding their offerings by looking at developer and software partnerships and adding new management capabilities and platform services. All of this gives buyers more options and thus assures that every path to justifying cloud services is explored fully.
Amazon Web Services Inc. has intimidated some cloud competitors with its strong early lead, but the early cloud market is a small fraction of the total cloud opportunity. More and more providers understand, however, that the full potential of the cloud is yet to be realized.
This recognition -- that the cloud is far from being locked up by early incumbents -- may bring about the most profound changes in 2013. The more cloud platform services we see, and the more cloud-specific applications are developed, the more total revenue the cloud can bring in. Ultimately, the cloud will be an instrument in the transformation of consumer behavior and worker productivity, and the seeds for both these revolutions will be sown in 2013 and be visible in services and revenues by this year's end.
About the author:
Tom Nolle is president of CIMI Corp., a strategic consulting firm specializing in telecom and data communications since 1982. He is the publisher of Netwatcher, a journal addressing advanced telecom strategy issues.