The latest trend in cloud technology involves businesses aggressively adopting numerous aspects of the cloud for greater operational efficiency, especially when it comes to enabling data centers with cloud technology. As a result, large enterprises are often left with the task of migrating hundreds of applications. Adopting a methodical approach towards classification and migration is one of the best approaches towards organizing applications into different categories. Afterwards, the appropriate deployment model can be identified for each application or groups of applications.

Where Do Applications Stand?

One of the first steps towards classification and migration involves analyzing the portfolio of applications from an application technology perspective. From here, enterprises can devise an appropriate deployment model consideration tailored for the portfolio at hand. In the realm of cloud enablement, this means viewing business applications from a technological point of view rather than viewing them from a functionality angle as would normally be done.

The key criteria for identifying workloads and applications to be migrated into the cloud involves determining whether an application can be viably run on a virtual machine stack and a platform that’s been commoditized and standardized.

Migration Methods

There are three distinct categories through which enterprises can accomplish the process of workload and application migration:

  • Lift and Shift – This process involves creating minor changes to workloads to accommodate a transition of network configurations, operating systems, and other foundation technology platform components. Lift and shift workload migration process normally operates within the realm of infrastructure as a service (IaaS), which gives enterprises the opportunity to run the workloads within a virtualized environment.
  • Transformation – This process involves transforming applications so that workloads can better operate within virtualized infrastructure and platform environments. This process of application migration also requires changes to the business logic to better adapt to the cloud environment and its various advantages. For instance, enterprises can include changes to workloads and applications that allow for improved information access through developed and externalized APIs.
  • Rebuild – This process is perhaps the most intensive of the three, since in involves making significant changes to the application in order to accommodate cloud technology. Not only does it involve significant changes to the business logic organization to enhance business capabilities, but it also involves ensuring that certain processes are able to be supported within the new cloud infrastructure, such as software libraries often used in cloud-bound applications.

Building a Framework

The next step in the classification and migration process involves enterprises deciding which deployment model to pursue. Enterprises can choose between public, private, and hybrid deployment models, but each choice offers its own sets of benefits and concessions in terms of agility, infrastructure, security, and regulatory needs.

Any deployment model created must be decided on a comprehensive deployment model that provides high-level decision-making criteria. IBM’s Cloud Affinity Analysis (CAT) application offers a structured approach towards migration assessment. CAT takes into account the potential business value versus cost of migration as well as the position of migratory applications in terms of key cloud characteristics.

Such considerations are represented in the form of cloud affinity as well as the overall pain and gain of migrating applications to the cloud:

  • Factors affecting “pain” include the amount and type of data transferred as well as security and compliance requirements associated with application workloads.
  • Factors affecting “gain” include the organization’s motivation towards migration as well as agility and elasticity requirements.

The above classification models and framework for migration and deployment should prove useful for enterprises interested in transitioning their applications to the cloud.