Our team here have prepared an overview of these steps several organizations took to move to cloud, based on the experiences of our service organization here at IBM Systems. The pathway would be very similar for other service providers too, so it's not something that's unique. The first step in the path to cloud, is to condition the existing infrastructure by planning for virtualized systems with a good service management interface. It's then the task of the decision-makers to define a cloud strategy. Without a strategy, cloud computing can be a threat to the CIO and the IT team. Threats like reduced control of IT services delivered over the Internet and cost differences between traditional and cloud-based IT, need to be addressed. However, with a strategy in place, cloud computing is a huge opportunity for the CIO. As we've seen, cloud lowers costs, enables a more responsive IT and paves the way for a more optimized delivery. With cloud, teams inside the company can concentrate on a greater range of services and capabilities that can be provided. Cloud helps get greater visibility in billing and charge-back to lines of business. It also lets administrators configure better security features to users' virtual environments. The next step is to analyze existing instances, another name for virtual machines in the cloud world, and choose an initial pilot project. The risks and migration costs may be too high for applications like Database, Transaction Processing, ERP and highly regulated instances. Certain instances like web servers, collaboration Infrastructure, deployment and test, and high-performance computing infrastructure, can be standardized. For the administrator, this means a wide variety of applications in the service catalog. If that isolated deployment was a success, it makes sense to extend and evolve more services into the catalog of your organization. You may start with, say, infrastructure as a service to give users in your organization the ability to access an operating system with a set of predefined resources and installed applications. If it makes more sense to move later to other service models that can also be considered as part of the evolution. Now, this slide shows the typical stack of a private cloud. Every layer has to be integrated seamlessly to create the features of the cloud. Software offerings predominantly rely on how platform and hardware components are provisioned and orchestrated. At the base we have service, also known as compute, then the storage controllers in fabric and the networking fabric. On top of that, we have the virtualization layer. The most popular hypervisors are shown here. VMware's ESXi is one popular hypervisor, and so is the open-source XenServer. We also have PowerVM, which you've worked on during this course. The POWER hypervisor is a firmware layer sitting between the hosted operating system and the server hardware. PowerVM is a combination of hardware, firmware and software that provides CPU network and disk virtualization. On top of that, we have the operating system layer. Red Hat Enterprise Linux is the world's most popular Enterprise Linux operating system. Finally, most organizations have a management layer, the out-of-band management, which performs tasks like firmer updates on the hardware servers and so on, like the hardware management console we have for IBM Power environments. These days, we have a cloud environment on top of that layer to manage multiple different vendors' infrastructures, right from this one cloud instance.
