1. Relocation This method can redeploy existing data and applications on cloud storage and computing resources without modification. This is usually the fastest and most direct migration method. It is not suitable for all types of applications, but it can work well with little cloud computing expertise or access to the underlying code base.
2. Refactoring This method modifies a small part of the underlying code base to optimize the workload and improve the reliability or performance in the public cloud. For example, the workload is modified to use the database services available in the cloud platform instead of the local database.
3. Enterprises can choose to modify the code of the workload a lot to use more cloud native services. This requires a clear and detailed understanding of the resources, services and infrastructure of the cloud provider. However, the overall characteristics and functions of the workload remain unchanged.
4. Reconstruction is usually the most complicated and demanding migration method. This requires recreating the workload from scratch to run efficiently in the cloud computing provider's environment. For example, you can use cloud native architectures such as microservices to redesign and rebuild aging legacy workloads.
5. Replacing this method means that the enterprise does not deploy, modify or recreate the workload, but uses the applications of third-party vendors, usually as SaaS products, instead of deploying, modifying or recreating the workload. The enterprise will only migrate the data of this application. For a simple example, it is easier to use a workload monitoring tool from a cloud computing provider than to try to deploy and use the same tool running in a local data center.