Lift and shift is a cloud migration strategy that involves moving existing applications and workloads from on-premises servers to a cloud environment with little or no modification. It's primarily used to reduce data center costs, improve scalability, and simplify infrastructure management. This method is also known as rehosting.
The core idea behind lift and shift is speed and simplicity. Instead of rewriting or refactoring applications, teams copy the entire system-operating system, databases, middleware, and configuration-into a cloud provider's virtual machine or container environment. The process typically uses tools like AWS Server Migration Service, Azure Migrate, or VMware vCenter Converter.
For example, a company running a legacy customer relationship management (CRM) system on physical servers in a local data center can use lift and shift to move it to Amazon EC2. The application continues to run as-is, with only network and storage configurations adjusted for the cloud. This reduces deployment time from months to days.
Lift and shift begins with discovery and assessment. Tools scan on-premises environments to inventory servers, applications, dependencies, and resource usage. This data determines which systems are suitable for migration and identifies potential risks like outdated software or unsupported APIs.
Once assessed, the migration phase begins. The source system is captured using disk imaging or virtualization tools. The image is then transferred to the cloud via dedicated network connections or secure file transfer protocols. After upload, the virtual machine is launched in the cloud with assigned IP addresses, security groups, and load balancers.
After deployment, the application is tested for functionality, performance, and security compliance. Monitoring tools track CPU, memory, and network usage to ensure stability. If issues arise, rollback procedures are triggered to restore the original state.
Lift and shift is most effective for applications that are stable, mission-critical, and not designed for microservices. It's ideal for organizations facing immediate pressure to reduce data center expenses or meet regulatory deadlines for cloud adoption.
| Use Case | Best For | Limitations |
|---|---|---|
| Legacy ERP systems | Quick cloud access, minimal downtime | High ongoing operational cost due to underutilized resources |
| Batch processing jobs | Scalable compute during peak hours | Not optimized for auto-scaling or serverless models |
| Development and testing environments | Fast provisioning, easy teardown | Often left running, increasing cloud spend |
A real-world example: a regional bank migrated its core loan processing system from on-site servers to Google Cloud Platform using lift and shift. The team spent seven days preparing the environment, five hours transferring the VM, and two days validating performance. Post-migration, they reduced monthly infrastructure costs by 42% and improved uptime from 97.6% to 99.9%.
Lift and shift is often misunderstood as a permanent solution. In reality, it's a transitional step. Applications moved via lift and shift may continue to consume high compute resources because they weren't designed for cloud efficiency. This leads to unnecessary spending-cloud providers charge based on usage, not hardware ownership.
Another misconception is that lift and shift eliminates technical debt. Moving an outdated monolithic app to the cloud doesn't fix poor code quality, tight coupling, or lack of monitoring. Without follow-up refactoring, teams risk creating "cloud sprawl"-an unmanaged collection of underused or duplicated VMs.
A key distinction: lift and shift is not the same as re-platforming. Re-platforming involves minor modifications-such as switching from a local database to a managed service like Amazon RDS-to improve performance and reduce maintenance. Lift and shift skips these changes entirely.
Don't use lift and shift for applications requiring real-time processing, frequent updates, or integration with AI/ML services. These systems benefit more from redesigns that adopt event-driven architectures or serverless functions.
Also avoid lift and shift if the application relies heavily on custom hardware, specific network configurations, or proprietary software not available in the cloud. Some government and healthcare systems face compliance barriers that prevent direct migration without approval from auditors.
I once observed a retail chain attempt lift and shift on a point-of-sale system tied to local barcode scanners. After migration, the system failed to recognize devices due to missing drivers. The team had to rebuild the application in a hybrid model, adding local edge devices. This added six weeks to the timeline and doubled the budget.
Start with a pilot project. Choose a non-critical application with clear metrics-like response time, error rate, and resource utilization-to measure success. Use automated tools to assess compatibility and estimate migration effort.
Set up tagging and cost allocation early. Assign metadata like department, owner, and purpose to each VM. This enables accurate tracking of cloud spend and prevents shadow IT.
Plan for optimization within 6-12 months. Monitor usage patterns and identify candidates for refactoring. Replace static VMs with auto-scaling groups or containerized services where appropriate.
What is lift and shift? It's a practical, low-risk way to begin cloud transformation. But it's not a long-term strategy. Success depends on treating it as a stepping stone-not a destination.