While technology and operations have evolved over the years, IT has always focused on a set of owned assets, with...
a set lifespan, that are managed by technical professionals. Cloud computing, however, is changing that. The cloud requires an IT operations team to embrace application populism -- or the fact that business departments, not just IT, are selecting and deploying new apps. In addition, cloud demands IT teams categorize resources on a sliding scale -- ranging from tactical to strategic -- while ensuring consistency for all applications.
The biggest change cloud presents to an IT operations team is that it empowers line organizations to undertake IT projects on their own. Software as a service (SaaS) and third-party integrators help make this possible. There are risks associated with this shift, especially around security, compliance and the integrity of workflows. As a result, IT organizations try to stop this practice -- but that's a mistake.
Business units will continue to deploy new cloud applications and IT should try to embrace that model. To do this, IT organizations must separate application review and application justification processes. If the cloud and new pricing models let departments justify their use of new applications, IT should accept those justifications.
Next, your IT operations team should focus on compliance and security reviews. These reviews should take two forms: a policy document that outlines requirements for a cloud application to be deployed, and a review process. To maintain that document and review process, IT teams must perform a regular survey of cloud application resources and determine what's needed to meet internal standards.
Cloud also changes IT operations because it introduces tactical resources -- or resources an organization rents for a given period of time -- into IT planning. IT is no longer dependent on capacity planning for central resources, tools and staff. The cloud lets IT address any temporary business needs in a more economical way.
To address this, your IT operations team should review past application proposals to see if the cloud provides a more economical option than data center technology did. Since SaaS and platform as a service (PaaS) clouds displace more costs than infrastructure as a service (IaaS), look hard at SaaS and PaaS options during this review. Keep careful records during the review process, as they're needed for the next step.
Next, revise application reviews and data center plans to reflect the availability of tactical resources. A good IT plan for cloud should visualize resources on a scale from very transient or tactical to long-lived and strategic. As resources move toward the strategic end of the scale, IT plans should specify how to put technical support in place and ensure rapid on board procedures. The final portion of the plan should size data center resources based on readily available cloud services. When establishing a resource scale, also consider the cost of security and governance for mission-critical applications.
Evolving application deployment models for cloud
The final hurdle when adapting IT operations for cloud is creating an application deployment model that works across that range of tactical and strategic resources. Applications often expand their scope with use, which may require a shift along the resource scale. Having different requirements for application deployment and support, based on the resources an application uses, makes deployment more difficult. Basic policies are needed to help users and IT cope with a growing inventory of resources. IT should consider three specific areas: enterprise architecture, DevOps and application lifecycle management (ALM).
A company is a web of interactions -- between people and IT -- that supports operations. IT connects these workflows, but as the cloud shifts application responsibility out to line departments, the risk of disconnecting these processes increases. Enterprise architecture involves the study of business processes and exchanges. It can identify business interactions not supported by IT workflows, and be a guide for information security and governance. Enterprise architecture becomes more important with cloud, so supporting it should be part of a cloud plan.
DevOps is a term used to describe automated deployment tools that build and sustain applications on their host platforms. Without these tools, organizations rely on expensive, manual processes. Some DevOps tools can be extended to a range of cloud platforms, but customization is beneficial. A DevOps strategy should support the scale of resources mentioned above.
ALM is the final step. The goal of ALM is to evolve applications by imposing specific testing and deployment rules, often through specialized ALM software. Many companies include security and governance validation in their ALM processes, and this is critical for cloud planning.
There are a number of steps to make an IT operations plan cloud-friendly, but none of them should create a barrier to cloud. A lack of attention to planning can cause issues that will discredit cloud. Plan now, and be successful instead.
IT teams require business know-how for cloud success
How cloud should transform business, IT mind-sets
Avoiding culture shock after a cloud deployment
New app development models emerge around cloud