The word cloud was everywhere in the high tech industry in 2010. The incredible rise of Amazon's public cloud offering and their success stories drew record interest from customers and technology providers alike. We saw everyone in the latter group start to "cloudify" their marketing. If you did not have a cloud strategy, you had the risk of falling behind. The race to "cloud" created a lot of confusion and there were very few, if any true cloud computing deployments beyond Amazon's success stories.
As enterprise early adopters wanted to bring these benefits to their infrastructure, they started looking under the covers of the generally available private cloud offerings on the market from startups and the established virtualization and management leaders and found that those vendors couldn't deliver on the promise of cloud computing because their solutions were not designed and built for cloud requirements and scale.
In 2011, I believe we'll see new innovative vendors deliver private cloud solutions built from the ground up to deliver cloud benefits to enterprise customers. This will help "clear" the skies and what were just trends, ideas or initiatives this year, will start becoming real and tangible in 2011. As a result, we'll see an acceleration of cloud computing deployment and usage beyond the current Web 2.0 world in traditional enterprise and service provider data centers.
This will be a perfect opportunity for IT to turn the tables and become an engine of innovation again. Cloud computing technologies will help with the management of data center infrastructure, which has become one of the top challenges in the enterprise, and in turn allow IT to focus on delivering new applications to the line of business. While virtualization has already brought a lot of efficiency gains to IT, there are still a number of missing pieces needed in order for IT be more agile and on the side of building competitive advantage rather than a cost center.
One of the first major steps in that direction will come from automation. Current virtualization implementations still require numerous manual steps and that is neither efficient nor scalable. Automation is the next logical step and eliminates human errors. Automation should start from the moment you start installing the infrastructure and allow it to scale up as fast as you can stack equipment. A few minutes of manual interaction per machine results in loss of efficiency and an increase in the potential for human errors as your infrastructure grows.
But automating the build up of infrastructure should only be the first step. As you manage the infrastructure and build applications on top of it, automation will keep gaining foothold so that tedious, error-prone but well understood processes can be achieved with the maximum efficiency possible.
And this efficiency will increasingly be achieved on commodity hardware and software. In 2006, Alessandro Perilli covered the launch of Amazon's Xen-powered virtual data center on demand, Amazon's public cloud offering, and highlighted that he had expected VMware to be the first to launch such a service and not Amazon. But Amazon innovated with a new approach and they were not the only ones doing so. Over the past years, giants like Google, Facebook and others have demonstrated that you can build and deliver world-class applications and services at very large scale without brand name hardware or expensive hypervisors.
This movement has started entering the enterprise world and I expect it to pick up momentum in 2011. As the price of the base hypervisor is rapidly declining with some being free all together, customers are more and more comfortable running various offerings in their data centers. One size does not fit all and one should use the hypervisor most suited to the use and application being built on top of it.
As organizations start using the same building blocks as the major public cloud providers, the move towards true hybrid clouds will become a greater reality in 2011. Public clouds have demonstrated the business benefits of cloud computing in terms of efficiency, scalability and agility. Those benefits can be achieved in great part on private infrastructure using private cloud offerings. IT can look to bring greater amounts of flexibility and agility behind their firewall and empower their internal business customers. But not every application will be required to run on the IT infrastructure and in some cases, the use of public cloud infrastructure will make more sense from an economic or architecture perspective.
This will create a co-existence model where IT can pick and chose which applications should run on their traditional core systems, which should run on a new breed of cloud enabled infrastructure behind the firewall and which should be moved to the public cloud. This hybrid model will allow an unprecedented level of elasticity.
Although initial interest in the cloud was primarily driven by cost savings, other aspects of the cloud promise have been picking up steam and I expect them to dominate the reasons for adoption and deployment through 2011. The level of innovation enabled by private and hybrid cloud technologies will allow IT to build and deliver better applications with virtually unlimited capacity, using third party resources when required. Moving beyond association with cost, IT will be associated with innovation again, bringing more competitive advantages to their organization.