VMware to Cloud Program Playbook
A phased program to migrate a VMware vSphere estate to public cloud, choosing per workload between hypervisor-compatible lift-and-shift and native replatforming, then optimizing and decommissioning vSphere.
VMware to Cloud Program Playbook
Many enterprises run hundreds of VMs on VMware vSphere. Rising licensing costs and datacenter pressure push them to the cloud. This program migrates a vSphere estate to public cloud, deciding per workload between a fast hypervisor-compatible lift-and-shift and a slower but cheaper native replatform.
Phase-by-Phase
Estate Assessment. Inventory every VM with its CPU, memory, storage, and dependency profile. Right-size aggressively; on-prem VMs are usually over-provisioned. For each workload choose a path: rehost onto a cloud VMware service (VMware Cloud on AWS, Azure VMware Solution, or Google Cloud VMware Engine) for speed, or replatform to native VMs and containers for cost.
Target Platform Build. Provision the target, whether a managed VMware SDDC or a native landing zone, configure networking and security, and set up replication tooling to stream VM data ahead of cutover.
Migration Waves. Migrate in dependency-grouped waves. Validate performance against the on-prem baseline before cutover, and keep a tested rollback. Hypervisor-compatible services let you move VMs without reformatting, which de-risks the early waves.
Optimize and Modernize. Lift-and-shift is the start, not the end. Right-size running workloads, replatform the candidates that justify it, and decommission vSphere to capture the licensing savings that motivated the program.
Team and Roles
An architect owns the path decision per workload. DevOps and SRE run the migration factory. DBAs handle data-heavy VMs. Security validates each wave. Product owns application acceptance.
Risks and Mitigations
Performance regression is mitigated by baseline benchmarking and validation gates. Licensing traps, especially carrying vSphere costs into the cloud, are mitigated by a clear decommission plan. Network latency between migrated and on-prem systems is mitigated by migrating dependency clusters together. Lift-and-shift cost creep is mitigated by the optimize phase and FinOps tagging.
Success Criteria
All in-scope VMs are migrated, run-rate cost drops below the vSphere baseline, performance reaches parity, and the vSphere estate is decommissioned.
Tooling
Use cloud migration and replication services, Terraform for the landing zone, Kubernetes for replatformed workloads, Datadog for cross-estate observability, Vault for secrets, and S3 for migration staging.