Lift and shift—also called rehosting—is one approach you can use to migrate applications such as data warehouses to the cloud. Rather than rearchitecting the application from the bottom up and making it cloud-native, you simply move the entire application to a cloud resource without impacting the associated business workflows, core logic, and data.
Lift and shift is often seen as an easier, faster way to move to a cloud data warehouse or other cloud-based infrastructure because it moves everything “as-is.” However, you may not be able to get the full benefit of the cloud without extra work. Let’s take a closer look at the benefits and challenges of a lift and shift approach, as well as how it stacks up against alternative cloud migration and modernization options.
Lift and shift opens a path toward IT modernization by moving resources to the cloud’s more scalable, flexible, and homogenous architecture. This ultimately reduces costs, improves resiliency, and performance. Because there is little to no redesign involved with lift-and-shift approaches, the transition is less complex and generally costs less than rearchitecting.
Top benefits of a lift and shift approach include:
Cost savings: You can take advantage of cloud’s economies of scale and eliminate on-premises data center infrastructure—both of which reduce costs
Accelerated digital modernization: With no changes required, you have an immediate path to the cloud and fewer chances for delays
High availability: Redundant cloud deployments support scalability and dependability. Apps and architecture stay essentially the same: you’re just rehosting your data on the cloud, and your business processes remain in place.
Lift and shift can offer an accelerated path toward cloud adoption, but it comes with its own set of challenges and disadvantages. Here are some downsides of the lift and shift approach:
Top benefits of a lift and shift approach include:
Missing out on cloud-native functionality: You may not be able to fully take advantage of cloud capabilities such as automated performance management, containerization, or ephemeral compute unless the application is rearchitected.
Old problems don’t magically disappear: If you don’t redesign legacy architecture problems and inefficiencies before migration, they will simply be replicated in the cloud environment—which means they’ll need the same amount of maintenance and upkeep once ported, and you’ll miss out on those cost savings.
Performance and latency challenges: Applications designed for on-premises environments may have problems running on cloud architecture without optimizations.
Risk of mapping failures: If the application requirements are not correctly mapped to the cloud environment, your migration could fail.
Staff learning curve: Organizations may lack the experienced cloud staff to operate the application in a cost-effective manner.
Deciding whether to lift and shift an application to the cloud is not a choice to be made lightly. You need to evaluate many critical components and interfaces to avoid disruptions. Here are three situations where a lift and shift approach may be the best option compared to other cloud migration strategies:
Organizations considering migrating to the cloud but concerned about cost and time to migrate. These organizations might have corporate mandates to make the shift but are constrained by both cost and time. Often these organizations plan on optimizing the applications once they have rehosted to the cloud.
IT teams looking to reduce on-premises infrastructure technical debt immediately. In this case, a team may need to move to the cloud fast to directly reduce costs associated with hardware, networking, and the staff required to maintain the infrastructure.
The application needs no changes—now or later. It is determined that the state of the target application does not need to change in the future and just needs to continue running without disruption. Also, the app will not lose operational efficiency in a move to the cloud.
Lift and shift may be a good strategy for moving certain workloads to the cloud as-is, but it won’t be the go-to option for all organizations. More organizations are looking for ways to leverage true cloud-native capabilities and benefits such as the use of containers, microservices, policy-driven resource utilization, and much more.
There are two main alternative migration approaches:
Refactoring/rearchitecting: With refactoring, the entire data warehouse or application (plus certain parts of its code) is modified and rearchitected for the cloud environment. The app may be adjusted for specific cloud services, as well. Fully rearchitecting a data warehouse or application for the cloud is resource- and time-intensive, but can yield significant gains over time: your architectures become more flexible and service-oriented, which allows you to take better advantage of cloud-based features and open up additional avenues of innovation.
Replatforming: Replatforming involves some aspects of lift and shift and some aspects of refactoring. In a replatform migration, the baseline application or analytics architecture is not significantly changed, but is fine-tuned before being ported to the cloud. This approach allows an organization to add a few modifications and new features, but start small and scale over time. You offload maintenance of the complex infrastructure to the cloud vendor, and development teams can continue to use their familiar resources.
With both refactoring and replatforming, it becomes much easier to leverage end-to-end cloud-native data management capabilities such as data integration and application integration, data quality, metadata management, governance, and more. These cloud-native capabilities give organizations an advantage as they look to modernize and future-proof their data management strategy.
Unlike the lift and shift approach, rearchitecting or replatforming analytics hubs such as data warehouses allows organizations to take full advantage of cloud infrastructure benefits including scalability, security, utility-based pricing, and robustness. They are also able to get better performance under high concurrency—a must-have in today’s competitive world where lines of business are demanding more real-time insights by more users for better decision making. Take a look at how Kelly Services used a rearchitecting approach and Informatica Intelligent Cloud Services to effectively build a modern data architecture in the cloud.
When planning your migration, take a close look at your goals: what specific benefits do you want to see after moving to the cloud? A fast and “easy” migration may sound good, but you may not be able to realize the full potential of the cloud without additional development work—something you don’t want to discover post-migration. And your data is too important to your business to risk making the wrong migration decision.
Watch our webinar on choosing the right path to cloud data warehousing to learn about the best data migration option for your organization. For more cloud migration and modernization resources, visit informatica.com/cloud-analytics.