Companies considering moving workloads to cloud environments five years ago questioned whether the economics of cloud were compelling enough. The bigger question at that time was whether the economics would force a tsunami of migration from legacy environments to the cloud world. Would it set up a huge industry, much like Y2K, of moving workloads from one environment to another very quickly? Or would it evolve more like the client-server movement that happened over 5 to 10 years? It’s important to understand the cloud migration strategy that is occurring today.
We now know the cloud migration did not happen like Y2K. Enterprises considered the risk and investment to move workloads as too great, given the cost-savings returns. Of course, there are always laggards or companies that choose not to adopt new technology, but enterprises now broadly accept both public and private cloud.
The strategy most companies adopt is to put new functionality into cloud environments, often public cloud. They do this by purchasing SaaS applications rather than traditional software, and they do their new development in a Platform-as-a-Service (PaaS) cloud environment. These make sense. They then build APIs or microservices layers that connect the legacy applications to the cloud applications.