Migration Planning Workflow

This skill defines the structured process for infrastructure migration planning. Use it for systematic, risk-managed migration execution.

Migration Planning Phases

Phase	Focus	Output
1. Discovery	Understand current state	Inventory document
2. Assessment	Evaluate migration options	Assessment report
3. Planning	Design migration approach	Migration plan
4. Preparation	Build target environment	Environment ready
5. Migration	Execute migration	Migrated workloads
6. Validation	Verify success	Validation report
7. Cutover	Switch production traffic	Cutover complete
8. Optimization	Tune and optimize	Optimized state

Phase 1: Discovery

Discovery Checklist

• Application inventory complete
• Infrastructure dependencies mapped
• Data flows documented
• Integration points identified
• Current costs documented
• Performance baselines captured
• Compliance requirements listed

Discovery Template

markdown

## Migration Discovery Report

**Project:** [migration name]
**Date:** YYYY-MM-DD
**Scope:** [on-prem to AWS / AWS to GCP / etc.]

### Application Inventory

| Application | Type | Language | Database | Traffic | Criticality |
|-------------|------|----------|----------|---------|-------------|
| api-service | API | Go | PostgreSQL | 1000 rps | Tier 1 |
| worker | Worker | Python | Redis | N/A | Tier 2 |
| frontend | SPA | React | N/A | 500 rps | Tier 2 |

### Infrastructure Inventory

| Resource | Current | Count | Utilization | Notes |
|----------|---------|-------|-------------|-------|
| Web servers | VM 4CPU/8GB | 10 | 40% avg | Auto-scaled |
| App servers | VM 8CPU/16GB | 20 | 60% avg | Fixed |
| Database | PostgreSQL | 1 primary + 2 replica | 70% | 500GB data |
| Cache | Redis | 3 node cluster | 30% | 10GB data |

### Dependencies

| From | To | Type | Migration Impact |
|------|-----|------|------------------|
| api-service | PostgreSQL | Direct | Must migrate together |
| api-service | Redis | Direct | Can migrate independently |
| frontend | api-service | HTTP | Can migrate independently |

### Data Inventory

| Data Store | Size | Growth Rate | Sensitivity | Retention |
|------------|------|-------------|-------------|-----------|
| PostgreSQL | 500GB | 10GB/month | PII | 7 years |
| S3 | 2TB | 100GB/month | Public | 1 year |
| Redis | 10GB | Stable | Session | N/A |

Phase 2: Assessment

Migration Strategy Options

Strategy	Description	Best For	Risk
Rehost (Lift & Shift)	Move as-is	Quick migration, legacy	Low
Replatform	Minor optimizations	Cloud-native benefits	Medium
Refactor	Redesign for cloud	Maximum benefit	High
Retire	Decommission	Unused systems	None
Retain	Keep in place	Not suitable for migration	None

Assessment Matrix

markdown

## Migration Assessment

### Per-Application Strategy

| Application | Strategy | Rationale | Effort | Risk |
|-------------|----------|-----------|--------|------|
| api-service | Replatform | Containerize, use managed DB | Medium | Low |
| worker | Rehost | Already containerized | Low | Low |
| legacy-app | Retire | Unused, no traffic | None | None |

### Migration Waves

| Wave | Applications | Dependencies | Duration | Risk |
|------|--------------|--------------|----------|------|
| 1 | worker, cache | None | 2 weeks | Low |
| 2 | api-service, DB | Worker | 4 weeks | Medium |
| 3 | frontend | API | 1 week | Low |

### Cost Analysis

| Item | Current Monthly | Target Monthly | Delta |
|------|-----------------|----------------|-------|
| Compute | $10,000 | $8,000 | -20% |
| Database | $5,000 | $6,000 | +20% |
| Network | $2,000 | $1,500 | -25% |
| **Total** | **$17,000** | **$15,500** | **-9%** |

### Migration Costs (One-time)

| Item | Cost | Notes |
|------|------|-------|
| Data transfer | $500 | S3 to S3 |
| Parallel running | $8,000 | 2 weeks overlap |
| Consulting | $10,000 | Optional |
| **Total** | **$18,500** | |

Phase 3: Planning

Migration Plan Template

markdown

## Migration Plan

**Project:** [name]
**Target Date:** YYYY-MM-DD
**Migration Lead:** [name]

### Timeline

| Phase | Start | End | Duration |
|-------|-------|-----|----------|
| Preparation | YYYY-MM-DD | YYYY-MM-DD | 2 weeks |
| Wave 1 | YYYY-MM-DD | YYYY-MM-DD | 2 weeks |
| Wave 2 | YYYY-MM-DD | YYYY-MM-DD | 4 weeks |
| Wave 3 | YYYY-MM-DD | YYYY-MM-DD | 1 week |
| Optimization | YYYY-MM-DD | YYYY-MM-DD | 2 weeks |

### Resource Requirements

| Role | Allocation | Duration |
|------|------------|----------|
| Migration Lead | 100% | Full project |
| Platform Engineer | 50% | Full project |
| Application Developers | 25% per team | Per wave |
| DBA | 50% | Waves 1-2 |

### Risk Register

| Risk | Likelihood | Impact | Mitigation |
|------|------------|--------|------------|
| Data loss | Low | Critical | Multiple backups, validation |
| Extended downtime | Medium | High | Rollback plan, parallel running |
| Performance degradation | Medium | Medium | Load testing, monitoring |
| Integration failures | Medium | High | Integration testing pre-cutover |

### Rollback Criteria

Migration will be rolled back if ANY of the following occur:
1. Data integrity issues detected
2. Performance degradation >20% vs baseline
3. Critical functionality not working
4. Security vulnerability discovered
5. Stakeholder veto during cutover window

Phase 4: Preparation

Environment Preparation

Preparation Checklist

markdown

## Environment Preparation Status

### Infrastructure

| Component | Status | Verified By | Date |
|-----------|--------|-------------|------|
| VPC/Network | Complete | @platform | YYYY-MM-DD |
| Compute clusters | Complete | @platform | YYYY-MM-DD |
| Database | Complete | @dba | YYYY-MM-DD |
| Cache | Complete | @platform | YYYY-MM-DD |
| Load balancers | Complete | @platform | YYYY-MM-DD |

### Connectivity

| Connection | Source | Target | Status |
|------------|--------|--------|--------|
| VPN tunnel | Source DC | Target VPC | Active |
| Database replication | Source DB | Target DB | Active |
| API connectivity | Source app | Target deps | Verified |

### Data Sync

| Data Store | Method | Lag | Status |
|------------|--------|-----|--------|
| PostgreSQL | Logical replication | <5 sec | Active |
| S3 | Cross-region replication | <15 min | Active |
| Redis | None (rebuild on migration) | N/A | N/A |

Phase 5: Migration Execution

Migration Execution Checklist

markdown

## Wave [N] Migration Runbook

**Date:** YYYY-MM-DD
**Start Time:** HH:MM UTC
**Migration Lead:** [name]

### Pre-Migration Checks

- [ ] All team members available
- [ ] Communication channels ready
- [ ] Rollback plan reviewed
- [ ] Stakeholders notified
- [ ] Monitoring dashboards open
- [ ] Data sync lag acceptable

### Migration Steps

| Step | Action | Owner | Expected Duration | Rollback |
|------|--------|-------|-------------------|----------|
| 1 | Freeze source changes | App team | 5 min | Unfreeze |
| 2 | Final data sync | DBA | 15 min | N/A |
| 3 | Stop source application | App team | 2 min | Restart source |
| 4 | Verify data sync complete | DBA | 5 min | N/A |
| 5 | Start target application | App team | 5 min | Stop target |
| 6 | Run smoke tests | QA | 10 min | Stop target |
| 7 | Update DNS/routing | Platform | 5 min | Revert DNS |
| 8 | Monitor for 30 min | All | 30 min | Rollback |

### Post-Migration Verification

- [ ] Application responding
- [ ] Error rates normal
- [ ] Latency within baseline
- [ ] Data integrity verified
- [ ] All integrations working

Phase 6: Validation

Validation Checklist

Validation Report Template

markdown

## Migration Validation Report

**Wave:** [N]
**Date:** YYYY-MM-DD
**Status:** [PASS / FAIL]

### Functional Validation

| Test | Expected | Actual | Status |
|------|----------|--------|--------|
| Login flow | Success | Success | PASS |
| API endpoints | All 200/201 | All 200/201 | PASS |
| Background jobs | Processing | Processing | PASS |

### Performance Validation

| Metric | Baseline | Target | Actual | Status |
|--------|----------|--------|--------|--------|
| p50 latency | 50ms | <60ms | 45ms | PASS |
| p99 latency | 200ms | <250ms | 180ms | PASS |
| Error rate | 0.1% | <0.2% | 0.08% | PASS |
| Throughput | 1000 rps | >900 rps | 1100 rps | PASS |

### Data Validation

| Check | Method | Result |
|-------|--------|--------|
| Row counts | SELECT COUNT(*) | Match |
| Checksums | MD5 on sample | Match |
| Foreign keys | Constraint check | Valid |
| Indexes | Index verification | Present |

Phase 7: Cutover

Cutover Criteria

ALL must be true to proceed with production cutover:

•All validation tests passed
•Rollback plan tested and ready
•Stakeholder approval obtained
•Communication sent
•On-call team ready
•Monitoring in place

Cutover Runbook

markdown

## Production Cutover Runbook

**Date:** YYYY-MM-DD
**Cutover Window:** HH:MM - HH:MM UTC
**Cutover Lead:** [name]

### Go/No-Go Checklist

- [ ] Validation report approved
- [ ] Rollback tested within 24h
- [ ] Stakeholder sign-off received
- [ ] Customer communication sent
- [ ] Team availability confirmed
- [ ] War room established

### Cutover Steps

1. **T-30min:** Final go/no-go meeting
2. **T-15min:** Final data sync verification
3. **T-0:** Begin cutover (follow wave runbook)
4. **T+30min:** Initial validation complete
5. **T+2h:** Extended monitoring
6. **T+24h:** Declare migration complete (or rollback)

### Rollback Decision Points

| Time | Trigger | Decision |
|------|---------|----------|
| T+15min | Critical errors | Immediate rollback |
| T+30min | >5% error rate | Rollback recommended |
| T+1h | Performance issues | Evaluate, possible rollback |
| T+2h | Stability issues | Extended evaluation |

Phase 8: Optimization

Post-Migration Optimization

• Right-size resources based on actual usage
• Enable cost optimization features
• Clean up source environment
• Update documentation
• Conduct lessons learned
• Archive migration artifacts

Optimization Template

markdown

## Post-Migration Optimization Report

**Migration:** [name]
**Optimization Period:** YYYY-MM-DD to YYYY-MM-DD

### Resource Optimization

| Resource | Initial Size | Optimized Size | Savings |
|----------|--------------|----------------|---------|
| Compute | 10 x m5.xlarge | 8 x m5.large | $400/mo |
| Database | db.r5.2xlarge | db.r5.xlarge | $200/mo |

### Source Environment Cleanup

| Resource | Action | Date | Savings |
|----------|--------|------|---------|
| Old VMs | Terminated | YYYY-MM-DD | $5,000/mo |
| Old storage | Deleted | YYYY-MM-DD | $500/mo |

### Lessons Learned

1. **What went well:** [items]
2. **What could improve:** [items]
3. **Action items for future migrations:** [items]

Anti-Rationalization Table

Rationalization	Why It's WRONG	Required Action
"Big bang is faster"	Big bang has higher risk	Migrate in waves
"Skip parallel running"	No safety net	Run parallel minimum 1 week
"Data validation is overkill"	Data loss is catastrophic	Validate every migration
"Rollback won't be needed"	Optimism != planning	Test rollback before cutover
"We can optimize later"	Later = never	Right-size during optimization phase

Pressure Resistance

User Says	Your Response
"Migrate everything this weekend"	"Big bang migrations carry unacceptable risk. Wave-based migration is required."
"Skip parallel running, too expensive"	"Parallel running is safety net. Cost is justified by risk reduction."
"Data validation takes too long"	"Data integrity is non-negotiable. Validation is required before cutover."
"We trust the process, skip rollback test"	"Rollback test is mandatory. Untested rollbacks fail when needed."

Dispatch Specialists

For migration planning tasks, dispatch:

code

Task tool:
  subagent_type: "ring:infrastructure-architect"
  model: "opus"
  prompt: |
    MIGRATION PLANNING REQUEST
    Type: [cloud-to-cloud / on-prem-to-cloud / upgrade]
    Source: [current environment]
    Target: [target environment]
    Scope: [applications/services]
    Timeline: [desired completion]
    Constraints: [budget, compliance, etc.]

For platform preparation:

code

Task tool:
  subagent_type: "ring:platform-engineer"
  model: "opus"
  prompt: |
    MIGRATION ENVIRONMENT PREPARATION
    Target Platform: [platform details]
    Services: [services to migrate]
    Requirements: [infrastructure requirements]