Cloud Architect
Cloud Migration Roadmap — From On-Prem to AWS Without Losing Your Mind
Key Takeaway
The Cloud Migration skill creates comprehensive migration plans — application assessment, dependency mapping, migration strategy per workload (rehost/refactor/rebuild/replace), timeline, risk mitigation, and cost projection. Stop migrating blind.
The Problem
"We need to move to the cloud." Six words that launch a 12-month project with a 60% failure rate.
Cloud migrations fail because:
- No assessment: "Let's just move the servers" (lift-and-shift everything, including the problems)
- Hidden dependencies: App A talks to App B through a hardcoded IP that nobody documented
- Cost surprises: Cloud costs 3x more than expected because nobody right-sized instances
- Downtime: Migration window was "1 hour" but took 14 hours because the database was 2TB
- Security gaps: On-prem firewall rules don't translate to security groups
- No rollback plan: If the migration fails at 3 AM, now what?
The Solution
The Cloud Migration skill generates complete migration plans with application assessment, the 6 R's strategy assignment (Rehost, Replatform, Refactor, Repurchase, Retire, Retain), dependency mapping, and phased execution timeline.
The Process
View details
You: Plan migration of our infrastructure to AWS:
- 15 application servers (mix of Java, Python, Node.js)
- 3 PostgreSQL databases (largest is 500GB)
- 2 Redis clusters
- 1 RabbitMQ broker
- NFS file storage (2TB)
- On-prem load balancer (F5)
- VPN connections to 3 partner networks
Current spend: $45K/month on colocation
The agent generates a comprehensive migration plan:
Phase 1: Assessment (Week 1-2)
- Application inventory with dependency matrix
- Database size and performance profile per DB
- Network flow analysis (what talks to what)
- Performance baselines (to compare post-migration)
Phase 2: Strategy Assignment
| Application | Strategy | Rationale | Target Service |
|---|---|---|---|
| Java monolith | Replatform | Too complex to refactor now | ECS Fargate |
| Python microservices (3) | Rehost → Refactor | Quick win, then optimize | EC2 → Lambda |
| Node.js APIs (4) | Rehost | Straightforward containerization | ECS Fargate |
| PostgreSQL primary | Replatform | Managed reduces ops burden | RDS Multi-AZ |
| Redis | Replace | Managed is cheaper + HA | ElastiCache |
| RabbitMQ | Replace | Managed equivalent exists | Amazon MQ |
| NFS storage | Replatform | EFS for shared storage | EFS |
| F5 load balancer | Replace | ALB is sufficient | ALB |
| Legacy reporting app | Retire | Nobody uses it (confirmed) | — |
Phase 3: Execution (6 waves over 12 weeks) Each wave: pre-migration validation → migrate → validate → monitor → sign-off
Phase 4: Optimization (post-migration)
- Right-sizing instances based on actual usage
- Reserved instances for predictable workloads
- Spot instances for batch processing
- S3 lifecycle policies for cold storage
Projected costs:
| Item | On-Prem | AWS (Month 1) | AWS (Optimized, Month 6) |
|---|---|---|---|
| Compute | $18K | $22K | $14K |
| Database | $12K | $8K | $6K |
| Storage | $5K | $3K | $2K |
| Network | $4K | $6K | $5K |
| Ops labor | $6K | $2K | $2K |
| Total | $45K | $41K | $29K |
The Results
| Metric | Wing-It Migration | AI-Planned Migration |
|---|---|---|
| Planning time | 2 weeks of meetings | 2 hours |
| Hidden dependencies found | During migration (panic) | During assessment (calm) |
| Cost surprise | +50-200% vs estimate | ±10% of projection |
| Downtime | Unplanned, extended | Planned, minimized |
| Rollback plan | "We'll figure it out" | Documented per wave |
| Success rate | ~40% on time/budget | ~85% |
Setup on MrChief
yamlShow code
skills:
- cloud-migration
- aws
- terraform # For IaC on the other side
- monitoring # For baseline + post-migration validation
Related case studies
Cloud Architect
Terraform Your Entire AWS Infrastructure — From Zero to Production
The Terraform skill generates complete Infrastructure as Code configurations — VPC, subnets, security groups, ECS/EKS clusters, RDS, S3, CloudFront, IAM — from a plain-text architecture description. Stop clicking through the AWS console.
SRE
Ansible Playbook for 50 Servers — Configure Everything in One Run
The Ansible skill generates complete playbooks for server configuration, application deployment, and infrastructure management. Describe what you need across your fleet, get idempotent, tested playbooks that configure 50 servers as easily as 1.
Backend Developer
API Design That Developers Actually Love — RESTful Done Right
The API Design skill generates complete RESTful API specifications — OpenAPI 3.1 schemas, endpoint design, authentication flows, pagination strategies, error handling, rate limiting, and versioning. Your agent designs APIs that follow industry best practices so your consumers don't hate you.
Want results like these?
Start free with your own AI team. No credit card required.