TH 11 ‐ Document different scenario & steps for migration - DevWithNehaCloud/Oracle-Cloud-Infrastructure-wiki GitHub Wiki
☁️ Cloud Migration – End-to-End Migration Lifecycle
📖 Overview
Cloud migration is the process of moving applications, databases, storage, networking, and business workloads from one environment to another while maintaining security, availability, performance, and business continuity.
Migration is not simply copying data from one location to another. It is a structured process involving assessment, planning, design, implementation, testing, validation, cutover, and post-migration support.
The primary goal is to migrate workloads with:
- Minimal downtime
- Minimal data loss
- Proper security controls
- Business validation
- Rollback capability if required
🌍 Types of Cloud Migration
1. On-Premises to Cloud
Move workloads from a physical data center to OCI.
On-Premises Data Center
│
▼
OCI
Common Reasons
- Hardware refresh
- Data center closure
- Cloud adoption
- AI / GPU requirements
- Scalability needs
2. Cloud to Cloud Migration
Move workloads between cloud providers.
OCI
│
▼
Azure
Common Reasons
- Multi-cloud strategy
- Business merger
- Vendor diversification
- Regulatory requirements
3. Region to Region Migration
Move workloads between OCI regions.
OCI Ashburn
│
▼
OCI Hyderabad
Common Reasons
- Disaster Recovery
- Latency optimization
- Regional expansion
4. Tenancy to Tenancy Migration
Move resources between OCI tenancies.
Source Tenancy
│
▼
Target Tenancy
Common Reasons
- Company split
- Ownership transfer
- Business restructuring
📦 Migration Strategies (6 R's)
Not every application is migrated in the same way.
| Strategy | Description |
|---|---|
| Rehost | Lift and Shift |
| Replatform | Minor cloud optimization |
| Refactor | Application redesign |
| Repurchase | Move to SaaS product |
| Retain | Keep existing workload |
| Retire | Decommission workload |
🚀 Cloud Migration Lifecycle
Phase 1 – Discovery & Assessment
Objective
Understand the current environment and identify everything that needs to be migrated.
Activities
Application Discovery
Collect:
- Application names
- Application owners
- Business criticality
- Dependencies
- Third-party integrations
Compute Discovery
Collect:
- Server Name
- Operating System
- CPU
- Memory
- Shape / Configuration
- Utilization
Storage Discovery
Collect:
- Block Volumes
- File Storage
- Object Storage
- Backup Storage
Database Discovery
Collect:
- Database Type
- Database Version
- Database Size
- Backup Requirements
Network Discovery
Collect:
- CIDR Ranges
- Subnets
- Route Tables
- Security Rules
- VPN Connectivity
- FastConnect Connectivity
Security Discovery
Collect:
- IAM Requirements
- Compliance Requirements
- Encryption Requirements
- Access Requirements
Deliverable
Migration Inventory Report
Client Sign-Off
Discovery Approval
💰 Phase 2 – Cost Assessment & Budget Approval
Objective
Estimate the cost of running workloads in the target cloud.
Activities
Calculate:
- Compute Cost
- Storage Cost
- Database Cost
- Network Cost
- Monitoring Cost
- Backup Cost
- Disaster Recovery Cost
Inventory Items Collected
- CPU
- Memory
- Shape
- Block Volume
- Object Storage
- Database
- Public IP
- Private IP
- Load Balancer
- Monitoring Requirements
Deliverable
Cost Estimation Sheet
Client Sign-Off
Budget Approval
Migration does not proceed until budget approval is received.
📐 Phase 3 – Architecture Design
Objective
Design the target cloud architecture.
Activities
Prepare:
- Region Selection
- Compartment Structure
- IAM Design
- Network Design
- Security Design
- Backup Design
- Monitoring Design
- DR Design
HLD (High Level Design)
Contains:
- Architecture Diagram
- Security Architecture
- Network Architecture
- DR Architecture
Focuses on:
WHAT are we building?
WHY are we building it?
LLD (Low Level Design)
Contains:
- IP Planning
- Security Rules
- Route Tables
- Server Configurations
- Build Instructions
Focuses on:
HOW exactly will it be built?
Deliverables
- HLD
- LLD
Client Sign-Off
Architecture Approval
🔗 Phase 4 – Dependency Mapping & Migration Waves
Objective
Understand application dependencies before migration.
Applications rarely work independently.
A single application may depend on:
- Databases
- APIs
- File Shares
- Active Directory
- DNS
- SSL Certificates
- Third-Party Services
Migration Waves
Applications are grouped into waves.
Wave 1 → Low Risk Applications
Wave 2 → Medium Risk Applications
Wave 3 → Business Critical Applications
Deliverable
Dependency Mapping Document
Client Sign-Off
Wave Plan Approval
🏗️ Phase 5 – Landing Zone Setup
Objective
Build the cloud foundation before application migration.
Activities
Create:
- Compartments
- Users
- Groups
- Policies
- Dynamic Groups
- VCN
- Subnets
- Route Tables
- Security Lists
- NSGs
- DRG
- Bastion
- Monitoring
- Logging
Typical Architecture
HUB VCN
│
┌───────────────┼───────────────┐
│ │ │
▼ ▼ ▼
SPOKE DEV SPOKE QA SPOKE PROD
Validation
Verify:
- IAM Access
- Connectivity
- Security Controls
Client Sign-Off
Landing Zone Approval
🌐 Phase 6 – Network Connectivity Setup
Objective
Connect source and target environments.
Options
On-Prem → OCI
- FastConnect
- Site-to-Site VPN
OCI → Azure
- FastConnect
- ExpressRoute
OCI Region → OCI Region
- Remote Peering Connection (RPC)
Validation
Perform:
- Ping Test
- Route Verification
- DNS Resolution
- Port Testing
Client Sign-Off
Connectivity Approval
🧑💻 Phase 7 – Development Environment Build
Objective
Create the first migrated environment.
Activities
Deploy:
- Application Servers
- Database Servers
- Storage
- Load Balancers
Configure:
- DNS
- Monitoring
- Security Rules
Validation
Development Team Testing
Sign-Off
Development Approval
📦 Phase 8 – Initial Data Migration
Objective
Move non-production data.
Common Methods
scp file.zip opc@target:/data
Other Methods:
- SCP
- Rsync
- Data Pump
- RMAN
- Object Storage
- Migration Tools
Important Note
Migration tools are generally used during data transfer stages.
The architecture, networking, security, and landing zone are usually built before migration tools are introduced.
Validation
Verify:
- File Count
- Record Count
- Checksums
- Data Integrity
Sign-Off
Initial Data Migration Approval
🧪 Phase 9 – QA Environment
Objective
Validate application functionality.
Activities
Perform:
- Functional Testing
- Regression Testing
- Integration Testing
- Performance Testing
Validation
QA Team Approval
Sign-Off
QA Sign-Off
👔 Phase 10 – User Acceptance Testing (UAT)
Objective
Business users validate the application.
Activities
Business teams execute:
- Login Validation
- Business Transactions
- Report Validation
- End-to-End Workflows
Deliverable
UAT Acceptance Letter
Sign-Off
Business Owner Approval
🔐 Phase 11 – Security Validation
Objective
Verify security posture before production.
Activities
Perform:
- Vulnerability Assessment
- Penetration Testing
- Compliance Review
- Security Audit
Validation
Security Team Approval
Sign-Off
Security Clearance
🔄 Phase 12 – Final Data Synchronization
Objective
Synchronize latest production data before cutover.
Activities
Perform:
- Delta Data Transfer
- Incremental Database Sync
- Final Backup
Validation
Compare:
- Source Records
- Target Records
- Database Counts
- Reports
Sign-Off
Migration Team Approval
🚀 Phase 13 – Production Cutover (Go-Live)
Objective
Switch production traffic to the target cloud.
Source Environment Activities
- Freeze Changes
- Stop Application
- Take Final Backup
Target Environment Activities
- Restore Latest Data
- Start Services
- Validate Application
Cutover Flow
Source Environment
│
▼
Final Sync
│
▼
Application Validation
│
▼
GO LIVE
🚦 Phase 14 – Go / No-Go Decision
GO Decision
Proceed when:
- Application is healthy
- Database is healthy
- Security team approves
- Business team approves
GO
NO-GO Decision
Rollback when:
- Critical defects exist
- Data validation fails
- Business validation fails
NO-GO
Stakeholders
- Cloud Team
- Network Team
- DBA Team
- Application Team
- Security Team
- Business Owner
↩️ Phase 15 – Rollback (Fallback) Plan
Objective
Return to the source environment if migration fails.
Common Rollback Triggers
- Data mismatch
- Critical application failure
- Performance degradation
- RTO violation
Rollback Activities
- Redirect Traffic
- Restore DNS
- Restart Source Environment
- Re-enable Original Services
Example
Rollback Window = 48 Hours
👀 Phase 16 – Hypercare Support
Objective
Provide intensive monitoring immediately after go-live.
Typical Duration
48 Hours to 2 Weeks
Activities
- 24x7 Monitoring
- Incident Management
- Performance Validation
- User Support
📚 Phase 17 – Project Closure & Handover
Objective
Transfer ownership to support teams.
Deliverables
- Inventory Report
- Cost Sheet
- HLD
- LLD
- Migration Report
- UAT Sign-Off
- Security Report
- Cutover Report
- Rollback Plan
Handover To
- Operations Team
- Monitoring Team
- Support Team
🌱 Post-Migration Optimization
After migration is completed:
Cost Optimization
- Right-size instances
- Remove unused resources
- Configure budgets
- Configure cost alerts
Monitoring
Monitor:
- CPU
- Memory
- Storage
- Application Health
- Database Health
Disaster Recovery
Validate:
- Backups
- Replication
- DR Procedures
- DR Drill
📈 Migration Success Criteria
Migration is considered successful when:
- All applications are operational
- No critical defects exist
- RTO targets are achieved
- RPO targets are achieved
- Security validation is passed
- Monitoring is active
- Backups are active
- Business owner signs off
- Legacy environment is safely decommissioned
🎯 Interview Summary
Cloud migration starts with discovery and inventory collection, followed by cost estimation, architecture design, dependency mapping, and landing zone setup. Connectivity is established between source and target environments, and workloads are deployed in Development, QA, and UAT environments for validation. After security testing and final data synchronization, production cutover is performed. A Go/No-Go decision is taken, and a rollback plan remains available until the migration is fully stabilized. Once the environment is stable, hypercare support, optimization, and project closure activities are completed.
📝 Industry Notes
1. Inter-Cloud Migration (OCI ↔ Azure ↔ AWS)
Inter-Cloud Migration is commonly used when organizations want to avoid vendor lock-in and take advantage of services offered by different cloud providers.
For example:
- AWS may provide better services for certain workloads.
- Azure may be preferred for Microsoft-based workloads.
- OCI may offer better pricing for Oracle workloads.
If a company depends entirely on a single cloud provider, future pricing, licensing, or service limitations may create operational risks. To reduce this dependency, organizations may migrate workloads between cloud providers or adopt a multi-cloud strategy.
Interview Note:
One common reason for Inter-Cloud Migration is avoiding vendor lock-in and leveraging the strengths of multiple cloud providers.
2. Inter-Region Migration (OCI Ashburn → OCI Phoenix)
Inter-Region Migration is commonly used for:
- Disaster Recovery (DR)
- Business Continuity
- Regional Expansion
- Regulatory Requirements
- Reducing latency for users in another geography
If a complete cloud region becomes unavailable because of a large-scale outage, natural disaster, or regional service disruption, workloads can continue running from another region.
Organizations maintain secondary regions and Disaster Recovery (DR) environments to ensure business continuity and minimize service disruption during regional outages. Critical workloads should not rely on a single region, as failures or large-scale incidents can impact availability.
Interview Note:
Inter-Region Migration is primarily performed for Disaster Recovery, Business Continuity, and reducing dependency on a single region.
3. On-Premises to Cloud Migration
Organizations migrate from on-premises environments to the cloud to avoid managing physical infrastructure.
In traditional data centers, companies must purchase:
- Servers
- Storage
- Memory
- Network Equipment
- Backup Hardware
During periods of high demand, hardware procurement can become expensive and time-consuming. Cloud platforms eliminate the need to purchase and maintain physical infrastructure because resources can be provisioned on demand.
This is one reason many organizations move from on-premises environments to cloud platforms.
Interview Note:
On-Premises to Cloud Migration helps organizations avoid hardware procurement challenges and provides scalability without purchasing physical infrastructure.
4. Inter-Tenancy Migration
Inter-Tenancy Migration is commonly used when resources must be moved between different OCI tenancies.
Typical scenarios include:
- Company mergers
- Company acquisitions
- Business unit separation
- Organizational restructuring
- Regulatory or ownership changes
Example:
A company acquires another company that already has its own OCI tenancy. To standardize operations, applications and resources may be migrated into a common tenancy.
Another example is when a business unit becomes an independent organization and requires its own OCI tenancy.
Interview Note:
Inter-Tenancy Migration is usually performed during mergers, acquisitions, ownership transfers, or organizational restructuring.