The 6 R’s of cloud migration: which strategy to use for each application
Migrating to the cloud does not mean applying the same strategy to all of the company’s systems. Each application has a different age, architecture, and importance to the business, and AWS organizes the possible migration approaches into six categories, known as the 6 R’s. Understanding each one helps plan a faster, cheaper, and lower-risk migration.
Rehost: move without changing
Rehost, also called lift and shift, is moving the application to the cloud practically without changes to the code or the architecture. It is the fastest strategy to apply and is usually the starting point for companies that need to leave the on-premises data center quickly, leaving deeper optimizations for a later stage.
Replatform: targeted adjustments to gain efficiency
Replatform involves making small adjustments to the application during the migration, without changing the core architecture. A common example is migrating a self-managed database to a managed service such as Amazon RDS, reducing operational effort without needing to rewrite the entire application.
Repurchase: swap for a ready-made solution
Repurchase means abandoning the current application and adopting an already existing solution, usually in the software-as-a-service model. It makes sense when maintaining and migrating the legacy system costs more than adopting a ready-made market alternative maintained by another vendor.
Refactor: rebuild to truly take advantage of the cloud
Refactor is the deepest strategy, and involves redesigning the application to take advantage of cloud-native resources, such as serverless architecture, containers, or microservices. It requires more time and investment, but usually brings the greatest gains in scalability, performance, and cost reduction in the long term.
Retain: keep it as is for now
Retain means consciously deciding not to migrate a specific application at the moment. It may be because the system is about to be discontinued, because there is a technical dependency too complex to solve now, or because the return of the migration does not justify the effort at that moment.
Retire: shut down what is no longer needed
Retire is identifying systems that no longer generate value for the business and simply deactivating them. It is common to find forgotten, duplicated, or replaced applications during the initial assessment of a migration, and shutting them down reduces cost and complexity even before the migration begins.
How to choose the right strategy for each application?
The choice depends on the importance of the application to the business, the technical complexity involved, the available budget, and the expected deadline for the migration. Critical systems with high usage volume usually justify a larger investment in refactor, while internal support systems can be resolved quickly with rehost or replatform.
In practice, most migrations combine several of these strategies at the same time, applying the right approach for each system, instead of using a single strategy for the company’s entire application portfolio.
Why does this planning matter?
Companies that migrate without defining the right strategy for each application usually spend more time and money than necessary, or lose the opportunity to come out of the migration with a more efficient architecture. A well-done technical assessment at the start avoids rework and speeds up the return on the cloud investment.
CloudDog has experience applying the different migration strategies for companies of all sizes on AWS. Learn about our AWS Cloud Migration service and plan the right strategy for each application in your company.
