Migrating to the cloud is not a single strategy but a portfolio of strategies applied workload by workload. The "7Rs" framework provides a structured way to evaluate each application and choose the right migration path based on business value, technical debt, and operational constraints.
The 7Rs Comparison Table
| Strategy | Description | Effort | Risk | Cost (Migration) | Cost (Ongoing) | Time | Best For |
|---|
| Rehost (Lift & Shift) | Move as-is to cloud VMs | Low | Low | Low | High (no optimization) | Weeks | Deadline-driven migrations |
| Replatform (Lift & Reshape) | Minor tweaks (e.g., managed DB) | Low-Med | Low-Med | Low-Med | Medium | Weeks-Months | Quick wins with managed services |
| Repurchase (Replace) | Switch to SaaS (e.g., Salesforce) | Medium | Medium | Medium (licenses) | Variable | Months | Commodity workloads (CRM, HR) |
| Refactor (Re-architect) | Rewrite for cloud-native | High | High | High | Low (optimized) | Months-Years | Strategic differentiators |
| Retire | Decommission the application | Low | Low | Savings | Savings | Weeks | Redundant/unused apps |
| Retain | Keep on-premises intentionally | None | None | None | Existing | N/A | Compliance, latency, near-EOL |
| Relocate | Move VMs at hypervisor level | Very Low | Low | Very Low | High | Days-Weeks | VMware-to-cloud (e.g., VMware Cloud on AWS) |
Migration Wave Planning Matrix
| Wave | Scope | Typical Apps | Strategy Mix | Duration | Prerequisites |
|---|
| Wave 0 | Foundation | Landing zone, networking, IAM | N/A (infra setup) | 4-8 weeks | Cloud account structure, compliance baseline |
| Wave 1 | Quick wins | Dev/test environments, static sites | Rehost, Replatform | 4-6 weeks | Landing zone complete |
| Wave 2 | Low complexity | Stateless web apps, batch jobs | Rehost, Replatform | 6-8 weeks | Monitoring & logging in place |
| Wave 3 | Medium complexity | Stateful apps, databases | Replatform, Refactor | 8-12 weeks | Data migration tools validated |
| Wave 4 | High complexity | Core business logic, legacy systems | Refactor, Retain | 12-24 weeks | Team upskilled, patterns proven |
| Wave 5 | Optimization | All migrated workloads | Right-sizing, reserved instances | Ongoing | Baseline cost data collected |
Migration Assessment Checklist
APPLICATION ASSESSMENT SCORECARD
================================
[ ] Business criticality (1-5) ___
[ ] Technical complexity (1-5) ___
[ ] Data sensitivity / compliance(1-5) ___
[ ] Integration dependencies (1-5) ___
[ ] Team cloud readiness (1-5) ___
[ ] Current operational cost $/mo ___
[ ] Expected cloud cost $/mo ___
[ ] Migration effort estimate weeks ___
[ ] Recommended strategy (7R) ___
[ ] Target wave ___
Migration Timeline Architecture
Month: 1 2 3 4 5 6 7 8 9 10 11 12
|----|----|----|----|----|----|----|----|----|----|----|
Wave 0 [=FOUNDATION==]
Wave 1 [==QUICK WINS==]
Wave 2 [===LOW COMPLEXITY====]
Wave 3 [====MEDIUM COMPLEXITY=====]
Wave 4 [=======HIGH COMPLEXITY========]
Wave 5 [===OPTIMIZATION==>
| | |
Discovery & Migration Optimization
Planning Execution & Governance
Common Migration Anti-Patterns
| Anti-Pattern | Problem | Better Approach |
|---|
| "Big bang" migration | All-or-nothing risk | Wave-based incremental approach |
| Lift & shift everything | No cloud benefits captured | Assess each app individually |
| Refactor everything | Paralysis, budget overrun | Refactor only strategic apps |
| Ignoring data gravity | Unexpected egress costs | Migrate data layer first |
| No rollback plan | Stuck if migration fails | Maintain hybrid connectivity |
| Skipping Wave 0 | Security and governance gaps | Invest in landing zone first |
Resources