With the growing move toward enterprise cloud adoption, and more broadly toward meeting the changing demands of IT consumers, IT organizations are grappling with the dual challenges of how to respond to these demands for change (i.e. what specifically to do) as well as how to carry out solutions that will best support these changes (i.e. how to manage, operate and support it).
The cloud manifests itself in a variety of forms, but regardless of sub-type, it is about being nimble and flexible. To support such a model, IT itself needs to transform in such a way that it too is nimble and flexible.
Customers often cite requirements, such as faster provisioning times, as examples of the improved agility realized when deploying a private cloud. Often when undertaking a cloud initiative, the effort is largely technology-focused. The emphasis thus becomes finding the right tool to be able to provision faster, as there is no question that automated provisioning and deprovisioning are important procedures for operating a private cloud. However, organizations that study the underlying causes for long provisioning times quickly realize that the actual deployment usually only accounts for a relatively small portion of the delay. An actual provisioning may be less than an hour, but the internal process flow to get to the provisioning may take weeks. So buying a tool, by itself, doesn’t adequately address the problem.
It turns out that a combination of IT policies, workflows and range of variability in solution choice tends to weigh far more heavily on end-to-end provisioning time than does the actual deployment. To be successful, IT itself needs to transform to a service provider model. Many organizations have begun to do this, and it’s time for more to follow suit.
An important first step in becoming a service provider is developing and defining services by attributes that end users value and presenting them in a service catalog. In addition, it’s critical to re-examine existing approval and change management workflows. Often, what is discovered is that much of the “overhead” that has been put in place is directly related to conditional choices with how a service is deployed. If the services become truly standardized, these processes can often become streamlined as well.
The two biggest challenges to this transformation are organizational structure and cost management. The traditional silos in IT made sense in a highly customized, physical world but need to be rethought to be able to meet the needs of a standardized, virtualized one. Cost management, too, must evolve to support a service-based model. Concepts like pay as you go are the hallmark of cloud services, and users are demanding this kind of flexibility.
Ultimately, we are talking about the transformation of IT into a service provider, or to put it another way, IT as a Service (ITaaS). As a full service provider, many IT organizations will ultimately need to be able to manage and support internal legacy environments, private cloud service delivery, as well as, in many cases, a variety of public services. Establishing a service provider mindset and capability is the only way to maintain the agility and flexibility required to respond to future demands.