In an era where business agility and technical resilience are more crucial than ever, organisations are increasingly embracing cloud migration as a strategic driver. This journey not only offers enhanced scalability, security, and cost efficiency but also propels businesses towards a future-proof infrastructure. Whether transitioning from on-premises to cloud, switching between cloud providers, or moving from cloud back to on-premises, the process is underpinned by critical considerations surrounding the “why,” “where,” and “how” of migration.

This article, from Senior Consultant Roxana Necula, delves into these questions, offering insights into the objectives, destinations, and strategies that shape successful cloud migration paths. Through exploring common motivations, comparing migration types, and evaluating the 5Rs framework (Rehost, Retire, Repurchase, Refactor, and Rearchitect), she lays out an introductory guide to navigating the complexities of cloud migration.

– Rajesh Jethwa, CTO, Digiterre

Introduction to Cloud Migration Strategies

We now see organisations more often go through cloud migration processes with their applications. That usually requires extensive planning and preparation and, in this process, the teams in charge usually address three important concerns:

  • the “WHY” – what are their objectives, business needs, specific requirements and consider their limitations
  • the “WHERE” (on-premises to cloud, cloud to cloud, or cloud to on-premises)
  • the “HOW” (one of the 5R cloud migration strategies: Rehost, Retire, Repurchase, Refactor, and Rearchitect)

The “Why”

The factors leading to the need for migrating can blend, or shift in importance over time for organisations, shaping how the process unfolds. There are many out there, but here are the most common reasons we encountered so far:

Why migrate from On-Premises to Cloud

  • Scalability and Flexibility: Cloud offers easy scalability based on demand.
  • Enhanced Security and Compliance: Cloud providers invest heavily in security measures.
  • Total Cost of Ownership Savings: Moving to a pay-as-you-go model can reduce total costs.
  • Handle fluctuating usage levels of the application (spikes of usage).

Why migrate from Cloud to Cloud

  • Operational Cost Optimisation: Different providers offer varying pricing models and discounts.
  • Service Integration: Migrating for better service or integration options.
  • Strategic considerations such as mergers and acquisitions or vendor lock-in avoidance.
  • Performance Optimisation: Switching between providers can improve performance.

Why migrate from Cloud to On-Premises

  • Regulatory Compliance: Some industries require data to be kept on-premises for compliance
  • Data Sovereignty: Data needs to remain within specific geographic boundaries or legal authorities.
  • Operational cost considerations: On-premises infrastructure may be more cost-effective for certain workloads.

The “Where”

The “WHY” leads to the “WHERE” to migrate: from on-premises to the cloud, between one cloud provider to another or take the infrastructure from the cloud back to on-premises. Here is a table of comparison between these types of migration, considering cost, scalability, reliability, performance, security, flexibility, and maintenance.

AspectOn-Premises to CloudCloud to CloudCloud to On-Premises
Total Cost of Ownership (TCO)  LowerModerateModerate
ScalabilityHigherHigherLower (depends on infrastructure capabilities of the on-premises)
ReliabilityModerateModerateModerate, but depends on the quality of on-premises hardware and maintenance
PerformanceHigher, with the ability to select optimal cloud resourcesHigherLower (depends on the quality of on-premises hardware and maintenance)
SecurityHigherHigherLower
FlexibilityHigherHigherLower
MaintenanceLowerLowerHigher

The “How”

Whether migrating from on-premises to cloud, between cloud providers or back to on-premises, there are several strategies available that serve as a guiding compass, such as the 5Rs framework introduced by Gartner (a global research and advisory company):  Rehost, Retire, Repurchase, Refactor and Rearchitect. The comparison below highlights the key differences between these strategies ordered by complexity and time to implement, including their descriptions and key concerns.

Migration StrategyRehost (lift and shift)RetireRepurchase (replace)Refactor (rearchitect)Rearchitect (rebuild)
DescriptionMove applications as-is to the cloud without making any changesDecommission or phase out applications that are no longer neededReplace existing applications with commercial off-the-shelf (COTS) solutions or SaaS offeringsOptimise applications for cloud environment without changing their core functionalityCompletely redesign and rebuild applications for cloud-native architecture
Applies to which migration typeOn-Premises to CloudAllOn-Premises to Cloud and Cloud to Cloud  All  All
ComplexityLowerLowerModerateModerateHigher
Time to ImplementFasterFaster  Faster  ModerateTime-consuming
Cost (CapEx)Lower  LowerHigherModerateHigher
Cost (OpEx)HigherLowerModerateModerate  Moderate  
FlexibilityLimitedN/ALimitedModerateHigher
PerformanceModerate (maintains legacy performance characteristics)N/AModerate (depends on the replacement solution)HigherHigher
ScalabilityModerateN/AModerate (depends on the replacement solution)ModerateHigher
RiskLowerLowerModerateModerateHigher

Examples of each strategy in practice

Rehost

  • lift and shift an on-premises web application to Amazon EC2 or Azure Virtual Machines

Retire 

  • decommission legacy applications no longer in use
  • decommission redundant infrastructure components after migrating to a cloud-based SaaS solution

Repurchase

  • replace on-premises file servers with Amazon S3 or Azure Blob Storage for scalable and durable object storage
  • replace custom-built event processing systems by Apache Kafka

Refactor

  • Refactor a monolithic application into microservices architecture

Rearchitect

  • rewrite applications using cloud-native frameworks like serverless (AWS Lambda or Azure Functions)
  • containerise applications for deployment on Kubernetes

Final thoughts on cloud migration

In the lifecycle of a long-lived application, it is common to traverse through multiple stages of migrations, either sequentially or concurrently, based on evolving business needs and technological advancements.

For instance, an application initially hosted on-premises may undergo the Rehost phase, where it is lifted and shifted to the cloud infrastructure without significant modifications. As the application matures in the cloud environment, organisations may opt to Refactor certain components to optimise performance or scalability. Subsequently, they might Revise specific functionalities to align with changing business requirements. Over time, the application might undergo a complete Rebuild to leverage cloud-native services for enhanced efficiency and innovation. Eventually, organisations might decide to Replace the application entirely or parts of it with a cloud-specific or newer solution.

This iterative journey exemplifies how applications can transition through multiple stages of migration, adapting to the evolving landscape of technology and business demands.

If you would like to know more about Cloud migration strategies visit our Services page or Contact us.

Follow Us

Get the latest news and stay up to date

Get in touch

If you would like to find out more, or want to discuss your current challenges with one of the team, please get in touch.