Workload by workload, based on data sensitivity, compliance obligations, performance profile and how predictably the cost needs to behave. Most enterprises end up with a mix. We operate AWS, Microsoft Azure, Google Cloud and governed private cloud, so the recommendation is based on fit rather than platform.

Planned waves, a tested rollback path for every cut-over and a parallel run where the workload demands it. We operate Epic EMR environments and core banking platforms for regulated banks, and our customers test failover on a regular schedule so the path is proven before it’s needed.

Jurisdiction, access and audit trail are settled in the target design before anything moves. Our environments are built for HIPAA, FCA/PRA and similar regimes, and the question of who can reach the data and from where has a documented answer in every environment we operate.

Close to the data they use, in an environment that meets the governance requirements around that data. For regulated or sensitive data that often means governed private cloud. For training, experimentation and burst capacity, public cloud is frequently the right fit. Migration is often the first step toward an environment that supports both. 

Scope sets the timeline. A single meaningful workload typically moves within weeks; a full core-system program runs months. We build the wave plan from the assessment and track delivery against it.