The Escape Plan
For a decade, “Hybrid Cloud” meant running VMware on-prem and VMware in the cloud. That era is ending. The dramatic pricing shifts from Broadcom have made “Native Cloud” not just an architectural goal, but a financial survival strategy.
Technical Deep Dive
1. The “Pet” to “Cattle” Shift
- VMware: You nurse sick VMs back to health. You use vMotion to keep them alive during maintenance.
- Native Cloud: You shoot sick instances and let the Auto Scaling Group replace them.
- Requirement: Your apps must handle sudden termination. Session state must be externalized (Redis/Memcached).
2. Right-Sizing (The ROI Maker)
- On-Prem: You provisioned for peak load + 20% buffer because adding RAM took weeks.
- Cloud: You provision for average load and auto-scale for peaks.
- Result: Most VMs can be downsized by 50% when moving to cloud, funding the migration cost.
3. Database Freedom
- Opportunity: Don’t just move Oracle/SQL Server on a VM to Oracle/SQL Server on EC2.
- Move: Switch to Amazon RDS or Azure SQL Managed Instance. Offload backups, patching, and HA to the vendor. This is where the real operational savings live.
How to Choose a VMware Migration Partner
If you need a massive datacenter exit: Kyndryl. They managed the infrastructure you are leaving; they know how to move it.
If you need a “Lift and Optimize” approach: Rackspace. They are great at moving VMs quickly and then optimizing them for cost.
If you need strategic financial modeling: Deloitte. They can build the business case to convince your CFO that the “Broadcom Tax” is worth leaving.
If you need application refactoring: SoftServe. They don’t just move VMs; they rewrite apps to be cloud-native (Kubernetes/Serverless).
Red flags:
- Vendors who suggest “VMware Cloud on AWS” (VMC) as a long-term solution (it’s still VMware licensing)
- No plan for “Right-Sizing” instances (moving 64GB on-prem VMs to 64GB cloud VMs is a waste)
- Ignoring the “Egress Cost” trap
- No automation strategy (Terraform/Ansible)
[!TIP] Not ready for Public Cloud? If you want to escape Broadcom pricing but keep your data on-premise, consider VMware to Nutanix as a private cloud alternative.
When to Hire VMware Migration Services
1. The Broadcom Renewal Shock
Your VMware renewal quote just arrived. It’s 4x what you paid last year. The CFO is furious.
Trigger: “We need to get off VMware NOW.”
2. Hardware End-of-Life
Your servers are 5 years old. You are facing a $2M capital expenditure (CapEx) to refresh the hardware.
Trigger: “Do we really want to buy more servers?“
3. Agility Blockers
Developers are waiting weeks for a VM. They want AWS/Azure APIs to provision resources instantly.
Trigger: “Shadow IT” (Developers using credit cards for AWS).
4. Datacenter Closure
Your lease on the colocation facility is expiring in 12 months. You don’t want to renew.
Trigger: Real estate consolidation.
5. Innovation Stagnation
You are spending 80% of your budget on “keeping the lights on” (patching vCenter) and 0% on AI/Innovation.
Trigger: “We are an IT shop, not a software company.”
Total Cost of Ownership: VMware vs Native Cloud
| Line Item | % of Total Budget | Example ($1M Project) |
|---|---|---|
| Migration Labor (Partners) | 30-40% | $300K-$400K |
| Cloud Infrastructure (Year 1) | 40-50% | $400K-$500K |
| Training (Cloud Skills) | 5-10% | $50K-$100K |
| Dual Run Costs (Overlap) | 10-15% | $100K-$150K |
Hidden Costs NOT Included:
- Egress Fees: Moving data out of the cloud is expensive.
- Software Licensing: Bring Your Own License (BYOL) vs Pay-As-You-Go (PAYG) for Windows/SQL.
Break-Even Analysis:
- Median Investment: $800K
- Annual Savings: $400K (Hardware + VMware Licensing + Power/Cooling)
- Break-Even: 2 years
VMware to Native Cloud Roadmap
Phase 1: Discovery & TCO Analysis (Months 1-2)
Activities:
- Run discovery tools (AWS Migration Evaluator / Azure Migrate)
- Map dependencies (App A talks to App B)
- Identify “Zombie VMs” (running but unused)
- Build the Business Case
Deliverables:
- TCO Report
- Migration Wave Plan
Phase 2: Landing Zone & Foundation (Months 3-4)
Activities:
- Build the Cloud Landing Zone (VPCs, Security Groups, IAM)
- Set up Hybrid Connectivity (Direct Connect / ExpressRoute)
- Establish FinOps tags and budgets
Deliverables:
- Secure Cloud Environment
- Network Connectivity
Phase 3: Migration Waves (Months 5-10)
Activities:
- Rehost (Lift & Shift): Move simple apps using tools (AWS MGN / Azure Migrate).
- Replatform (Lift & Reshape): Move DBs to RDS, Web Apps to App Service.
- Refactor (Rewrite): Rewrite critical apps to Serverless/Containers.
Deliverables:
- Migrated Workloads
- Decommissioned On-Prem Hosts
Phase 4: Optimization (Months 11-12)
Activities:
- Right-Sizing: Downsize instances based on actual cloud usage.
- Reserved Instances: Commit to 1-3 year plans for 40% savings.
- Modernization: Start refactoring the “Lifted & Shifted” apps.
Deliverables:
- Optimized Cloud Bill
- Fully Retired Datacenter
Architecture Transformation
graph TD
subgraph "Legacy VMware"
A["Load Balancer (F5)"] --> B[Web VMs]
B --> C[App VMs]
C --> D["DB VMs (SQL/Oracle)"]
E[vCenter] --> B
end
subgraph "Native Cloud"
F["Cloud Load Balancer (ALB)"] --> G[Auto Scaling Group]
G --> H["Container Service (EKS/AKS)"]
H --> I["Managed DB (RDS/SQL MI)"]
J[Infrastructure as Code] --> G
end
style E fill:#f9f,stroke:#333,stroke-width:2px
style J fill:#bbf,stroke:#333,stroke-width:2pxPost-Migration: Best Practices
Months 1-3: FinOps
- Tagging: Ensure every resource has an “Owner” and “Cost Center” tag.
- Budgets: Set strict budgets and alerts. Cloud costs can spiral if unchecked.
Months 4-6: Automation
- IaC: Stop clicking in the console. Move to Terraform or Pulumi for all infrastructure changes.
- CI/CD: Automate deployments. No more manual file copies.
Expanded FAQs
Why not just use VMware Cloud on AWS (VMC)?
Answer: VMC is great for speed (no refactoring), but you still pay the “VMware Tax” (licensing) plus the AWS infrastructure cost. It is often more expensive than on-prem. Native cloud removes the VMware licensing cost entirely.
How do we handle IP addresses?
Answer: You don’t. In the cloud, IP addresses are ephemeral. You must switch to DNS-based service discovery. Hardcoded IPs are the #1 cause of migration failures.
What about our legacy OS (Windows 2008)?
Answer: Cloud providers offer “Extended Security Updates” (ESU) if you move to their platform, but you should prioritize upgrading. Running EOL operating systems is a massive security risk.
Is “Lift and Shift” bad?
Answer: No, it’s a valid first step. It gets you out of the datacenter quickly. But if you stop there, you will pay more than on-prem. You must have a Phase 2 plan to optimize and modernize.
How do we move 500TB of data?
Answer: For massive data, network transfer is too slow. Use physical devices like AWS Snowball or Azure Data Box to ship disks to the cloud provider.