6 Rs of Cloud Migration: What It Is and When to Use It

Definition

The 6 Rs of Cloud Migration is a framework promoted by AWS that provides six common strategies for migrating applications and infrastructure from an on-premises environment to the cloud. [11, 18] These strategies—Rehost, Replatform, Repurchase, Refactor, Retire, and Retain—offer a vocabulary for categorizing every component in an IT portfolio, helping organizations to plan their migration approach based on business drivers, technical requirements, and resource constraints. [7] The framework helps solve the complex problem of how to approach a large-scale migration by breaking it down into a series of well-understood patterns rather than treating every application as a unique challenge.

How It Works

An effective cloud migration begins with a discovery phase where an organization assesses its application portfolio. [3] During this assessment, each application or workload is mapped to one of the migration strategies. Over time, AWS has expanded the framework to include a seventh strategy, Relocate, making it the "7 Rs". [4, 5, 10] The choice of strategy for each application is a trade-off between the speed of migration and the degree of cloud optimization achieved. [1]

Here are the 7 common strategies as of 2026:

1. Rehost (Lift and Shift)

This is the most common and typically fastest migration strategy. [2] It involves moving applications from the source environment to the AWS Cloud with minimal or no changes. [9, 10] The application is essentially lifted from its on-premises server and shifted to a virtual server, such as an Amazon EC2 (Elastic Compute Cloud) instance. This approach is often automated using tools like AWS Application Migration Service (AWS MGN). [14]

Effort: Low. The primary work involves replication and cutover.
Benefit: Speed. Ideal for large-scale migrations or when facing a hard deadline, like a data center lease expiring. [13]
Drawback: You may not immediately leverage cloud-native benefits like auto-scaling or managed services, potentially carrying over existing performance or architectural issues. [7]

2. Replatform (Lift, Tinker, and Shift)

Replatforming involves making a few targeted cloud optimizations to achieve a tangible benefit without changing the application's core architecture. [9, 17] This might involve migrating an on-premises, self-managed Oracle or SQL Server database to a managed service like Amazon RDS (Relational Database Service) or moving an application to a managed platform like AWS Elastic Beanstalk. [14]

Effort: Low to Medium. It requires more planning than rehosting but avoids the complexity of a full re-architecture.
Benefit: Achieves a good balance of speed and cloud optimization, improving performance, cost, or operational resilience. [1]
Drawback: Can introduce a small amount of development work and requires careful testing to ensure compatibility.

3. Repurchase (Drop and Shop)

This strategy involves moving to a different product, often by ending a traditional license and repurchasing a Software-as-a-Service (SaaS) solution. [11, 14] Examples include moving from a self-hosted CRM to Salesforce, or from an on-premises email server to a cloud-based email service. The focus is on switching what you use, not just where it runs.

Effort: Varies. The primary effort is in data migration, user training, and integration with existing systems.
Benefit: Reduces the burden of managing underlying infrastructure and software, shifting it to the SaaS provider. [18]
Drawback: Potential for vendor lock-in, data migration complexity, and loss of custom features from the legacy application.

4. Refactor / Re-architect

Refactoring is the most intensive strategy and involves fundamentally re-imagining how an application is architected and developed to be fully cloud-native. [4, 9] This often means breaking down a monolithic application into microservices, leveraging serverless technologies like AWS Lambda and event-driven architectures, or using managed container orchestrators like Amazon EKS (Elastic Kubernetes Service). [12]

Effort: High. This is a significant development project.
Benefit: Maximizes the benefits of the cloud, including agility, scalability, performance, and cost-effectiveness at scale. [1, 18]
Drawback: The most time-consuming and expensive strategy upfront. AWS often advises against refactoring during a large-scale migration, recommending instead to rehost or replatform first and then modernize later. [10]

5. Retire

During the discovery phase, organizations often find that a percentage of their IT portfolio (typically 10-20%) is no longer useful and can be decommissioned. [3, 11] The retire strategy is simply the process of identifying these applications and turning them off.

Effort: Low. The main work is analysis to ensure no other critical systems depend on the application being retired. [11]
Benefit: Immediate cost savings in infrastructure, maintenance, and licensing. It also reduces the attack surface and the overall complexity of the migration.
Drawback: Requires careful dependency mapping to avoid accidentally breaking a business process.

6. Retain

This strategy means leaving certain applications in the source environment. [7] There are many valid reasons for this: some applications may be subject to strict regulatory or data residency requirements, others may be too complex or costly to move, or a migration may simply not be a business priority at the time. [14] These applications are often revisited at a later date.

Effort: None (in the short term).
Benefit: Avoids the cost and risk of migrating unsuitable or low-priority applications.
Drawback: The organization does not gain any cloud benefits for these applications and must continue to maintain the on-premises infrastructure.

7. Relocate

This is a newer strategy added by AWS, primarily focused on moving workloads between hypervisors with minimal changes. [5, 10] The most common use case is relocating virtual machines (VMs) from an on-premises VMware environment to VMware Cloud on AWS. This allows organizations to continue using their existing VMware management tools and skillsets while running on AWS infrastructure.

Effort: Low. Similar to rehosting but specific to hypervisor-level moves.
Benefit: Very fast for VMware-based workloads and allows for a consistent hybrid cloud operating model.
Drawback: Highly specific to certain technologies like VMware and may not offer deep integration with other AWS services initially.

Key Features and Limits

As a conceptual framework, the 6 Rs (or 7 Rs) do not have technical limits like service quotas. Instead, the key considerations involve strategic trade-offs:

Migration Velocity: Rehost and Relocate offer the fastest path to the cloud, while Refactor is the slowest.
Cost of Migration: The cost generally increases with the level of modification. Rehost is the cheapest upfront, while Refactor is the most expensive.
Long-Term Operational Cost: A successful Refactor or Replatform can lead to the lowest long-term operational costs by leveraging cloud efficiencies. A Rehost may carry over on-premises inefficiencies, leading to higher costs over time.
Business Agility: Refactoring provides the greatest potential for increased business agility by enabling faster feature releases and innovation.
Resource Availability: The choice of strategy depends heavily on the availability of skilled personnel, budget, and time. [1]

Common Use Cases

Rehost: An organization needs to exit a data center quickly due to a lease expiration. They use AWS Application Migration Service to lift-and-shift hundreds of servers to Amazon EC2 instances.
Replatform: A company's e-commerce site runs on a self-managed MySQL database on a physical server. As part of the migration, they move the application to EC2 but migrate the database to an Amazon RDS for MySQL Multi-AZ instance to improve reliability and reduce administrative overhead.
Repurchase: A business is running an outdated, on-premises version of a third-party CRM system. Instead of migrating it, they move to a SaaS-based CRM from the AWS Marketplace, eliminating the need to manage the software and its infrastructure.
Refactor: A monolithic application for processing financial transactions is becoming a bottleneck, making it difficult to add new features. The development team re-architects it into a set of microservices running on AWS Lambda, communicating via Amazon SQS (Simple Queue Service) and Amazon EventBridge, to improve scalability and development speed.
Retire: During a portfolio analysis, a company discovers three legacy reporting servers that have not been accessed in over a year. The data has been archived, and the servers are safely decommissioned, saving licensing and maintenance costs.

Pricing Model

The 6 Rs framework itself is a concept and has no direct cost. However, the chosen strategy directly impacts the cost of migration and subsequent AWS bills:

Rehost/Relocate: Costs are often similar to on-premises but shift from CapEx (Capital Expenditure) to OpEx (Operational Expenditure). You pay for Amazon EC2 instances, Amazon EBS (Elastic Block Store) volumes, and data transfer. Without right-sizing, this can be more expensive than anticipated.
Replatform: Pricing is similar to rehosting, but you may pay for managed services like Amazon RDS. While the service itself has a cost, it can reduce operational labor costs, leading to a lower Total Cost of Ownership (TCO).
Repurchase: The cost model shifts entirely to the SaaS provider's subscription fees.
Refactor: This strategy can lead to the most optimized cloud spend. Using serverless services like AWS Lambda means you pay only for the compute time you consume, down to the millisecond, with no charge for idle time. However, it requires a significant upfront investment in development.
Retire/Retain: Retire directly reduces costs. Retain has no impact on AWS costs but means on-premises costs continue.

To estimate the cost of any migration strategy, use the AWS Pricing Calculator.

Pros and Cons

Pros:

Structured Approach: Provides a clear, well-defined set of options that simplifies migration planning for large, complex portfolios.
Common Language: Establishes a shared vocabulary for business and technical stakeholders to discuss migration plans.
Flexibility: It is not an all-or-nothing framework; organizations can and should apply different strategies to different applications. [13]
Prioritization: Helps organizations prioritize migration efforts based on business value, complexity, and urgency.

Cons:

Oversimplification: The 'Rs' are high-level categories. The actual implementation of any strategy, especially Replatform and Refactor, involves many detailed architectural decisions.
Risk of Poor Choices: Without a thorough assessment, choosing the wrong strategy (e.g., rehosting an application that desperately needs to be refactored) can lead to technical debt and higher long-term costs.

Comparison with Alternatives

The AWS 6 Rs framework is an evolution of Gartner's original "5 Rs" of cloud migration (Rehost, Refactor, Revise, Rebuild, Replace). [5, 11, 15] AWS adapted and expanded this model, and it has since become the de-facto industry standard for discussing migration strategies. While other cloud providers have similar concepts, the AWS 7 Rs are the most widely recognized and referenced framework in the context of cloud migration planning.

Exam Relevance

The 6 (or 7) Rs are a fundamental concept for several AWS certifications and are frequently tested in scenario-based questions.

AWS Certified Cloud Practitioner (CLF-C02): Candidates should be able to define each of the 6 Rs at a high level.
AWS Certified Solutions Architect - Associate (SAA-C03): This is a core topic. Exam questions often present a business problem (e.g., "a company needs to migrate quickly with minimal changes") and require the candidate to select the appropriate 'R' (in this case, Rehost). [19, 22] You must understand the trade-offs between each strategy.
AWS Certified Solutions Architect - Professional (SAP-C02): Expect more complex scenarios involving hybrid architectures, multi-application migrations, and the long-term strategic implications of choosing one 'R' over another. [21]
AWS Certified Migration - Specialty (retired): This was the central theme of the entire exam.

Examinees must know when to apply each strategy based on constraints like cost, time, performance requirements, and operational overhead.

Frequently Asked Questions

Q: Which 'R' is the most common migration strategy?

A: Rehosting (Lift and Shift) is often the most common strategy, especially in the initial phases of a large-scale migration. [2] Many organizations rehost a majority of their applications to exit a data center quickly and then follow up with optimization and modernization (replatforming or refactoring) once the applications are running in AWS.

Q: How do I choose the right strategy for my application?

A: The choice depends on a thorough assessment of both business goals and technical characteristics. Key questions to ask include: What is the business driver for moving this application (e.g., cost savings, agility, reliability)? What is the application's architecture? Is it tightly coupled with other systems? What is our team's skill set? What is our timeline and budget? [1] Tools like the AWS Application Discovery Service can help map dependencies to inform these decisions. [3]

Q: Can I use multiple strategies in a single migration project?

A: Yes, absolutely. It is a best practice to apply different strategies to different applications within the same portfolio. [13] A typical large-scale migration will involve a mix of all the Rs: retiring obsolete applications, retaining those not ready to move, rehosting the bulk of the portfolio, replatforming key databases, and refactoring the most critical, high-growth applications.

This article reflects AWS features and pricing as of 2026. AWS services evolve rapidly — always verify against the official AWS documentation before making production decisions.