While many enterprise CIOs have “moving to the cloud” as part of their vision, a lot of clients we speak with are still determining their strategy for what to move to the cloud. When thinking about the public cloud, I often recommend a “toe in the water” approach. This approach will allow you to take advantage of the strengths of public clouds—low costs and elasticity, for example—while avoiding their weaknesses, mainly the lack of  enterprise-friendly SLAs (which we discuss in the Seek services, not just servers post).

Here are a few example use cases that are helpful in testing a public cloud:

• Customer service trials – One of our clients used the Windows Azure platform to test out a new service offering. This was a great way to determine the viability of the new offering without having to invest in the infrastructure to support it.
  
• Periodic data crunching – Many organizations are tasked with running reports that use a vast amount of data, which, in turn, requires a vast amount of processing power. Using the scalability of the public cloud is a great way to economically access needed processing power on a quarterly or monthly basis.
  
• Non-sensitive enterprise applications – For applications that do not contain sensitive information, using the public cloud is a good way to lower costs and access needed capacity. In addition, Microsoft’s hybrid model (leveraging, for instance, Windows Azure AppFabric) enables scenarios such as putting a front-end application in the cloud while keeping sensitive, back-end data or processing on premise. 

For many established enterprises, trials such as these are important because it gives them a taste for how to leverage the cloud, train IT staff and slowly introduce the shift from CAPEX to OPEX into the business culture. After all, for many large IT departments, moving to the cloud represents a big transition in culture. This is why CGI’s cloud consulting practice spends a lot of time working with clients on establishing a roadmap and running test cloud cases. As cloud computing evolves, their enterprises will be able to evolve right along with it.