Lift & Shift Landing Zone

⭐️⭐️☁️ PlatformA dedicated landing zone optimized for lift & shift workloads enables quick onboarding and efficient operations.

While lift & shift may not the be first thing on the cloud foundation team’s mind when thinking about their organization’s cloud journey, lift & shift is a valid and important complement to cloud-native applications in a cloud adoption journey. A dedicated lift & shift landing zone can help onboarding these workloads quickly and without too much organizational pain.

Before you get started however, it’s paramount that your organization’s cloud strategy clearly lays out the motivation and expected results for lift & shift migrations. Your cloud strategy should at least answer the following questions

  • Do you need to vacate an on-prem datacenter or MSP by a set date?

  • Do you plan on leveraging lift & shift to quickly ramp up cloud usage to meet negotiated cloud spend commitments?

  • Do you plan to simply re-host lift & shift workloads or do you plan to later re-platform or even refactor the same workload?

  • What know-how exists within the organization to build and operate servers? What cloud skills are available?

  • Who is going to operate applications? Are divisions going to be responsible for operations themselves and demand self-service or is a central IT team going to take care of applications for them?

Equipped with answers to above questions, the cloud foundation team can start designing and building a lift & shift landing zone.

Best Practices for Designing and Building a Lift & Shift Landing Zone

An important pre-requisite for establishing a successful lift & shift landing zone is going through the process of Shared Responsibility Model Alignment with all cloud foundation stakeholders. From our experience, it’s useful to start by reviewing the cloud strategy and selecting the primary approach to the landing zone design.

Lift & Shift Landing Zone Implementation Approaches

Dedicated Lift & Shift Cloud Tenants

When the organization plans on re-hosting only as an intermediate step before re-platforming or even refactoring workloads, designing the landing zone to provide dedicated cloud-tenants like an Azure Resource Group for every application teams offers several advantages.

From our experience this approach works best for cloud foundation teams that already have (or plan) strong investments into Cloud-native Landing Zone capabilities. One downside is of course that building the required cloud proficiency with application teams takes time and may limit cloud adoption velocity. Cloud foundation teams should thus consider how to establish a low barrier of entry and reasonable “learning curve” in Shared Responsibility Model Alignment.

VMware on AWS, VMware on Azure Etc.

Many organizations have existing investment into VMware based server provisioning and operational know-how. Cloud provider services like VMware on AWS or VMware on Azure allow organizations to quickly transfer these skills to the cloud without major adaptions or interruptions to existing workflows.

This lift & shift approach is a good choice when the organization plans on simply re-hosting workloads as quickly as possible to meet spend commitments or vacate on-premise datacenters. This approach can also drastically simplify technical migration of existing machines because it’s part of an integrated solution.

The downsides of this approach are that it offers little to no opportunity for building cloud skills in the organization and limits migration options to strict re-hosting.

Fully Managed Virtual Servers

In this model, a “server platform team” provides fully managed virtual servers to application teams, for example using existing ITSM processes already established in the organization. The server platform team manages the deployed machines in a separate part of the Resource Hierarchy. The server platform hides away all complexity of server deployment from application teams, meaning that they never access the “cloud”, but only receive access to their managed servers. This approach is useful when application teams lack cloud proficiency (as well as the ambition to acquire it) and re-hosting workloads is all that’s desired.

One downside of this approach is that delivering this level of service requires significant investment of time and resources for the server platform team. Especially considering that the team takes on a lot of responsibility for operating the deployed servers as well. From our experience we have seen cloud foundations have success with this approach when they can form a dedicated server platform team around members with existing on-premise server operations know-how who are also eager to acquire cloud skills. These are essential for building capabilities like Shared VM Image Repository and SOC Integration as well as strong expertise in cloud automation. The cloud foundation team on the other hand can support the server platform team by setting up essential services like a Virtual Network Service and establishing a Foundation Service Platform.

Even though virtual machines are the meat and bones of a lift & shift landing zone, the cloud foundation team should also consider how it will deliver related IT infrastructure services. Most application teams will also require Managed DNS Services and Managed SSL Certificates for their workloads, as well as a Managed bastion hosts to access their virtual machines. In many organizations, these capabilities already exist for on-premise managed virtual machines and can be easily transferred to the cloud.

An important caveat however is that many of these existing processes will be ISMS based and eschew cloud-native practices like API-driven automation. Re-implementing these capabilities (very often based on top of the services provided by a cloud provider) can offer significant efficiency improvements for application teams capable and willing to take advantage of them.

Leverage Cost Optimization Opportunities

Unfortunately “always-on” Lift & Shift workloads are the anti-thesis to the cloud’s on-demand pay-per-use pricing model. Cloud foundation teams should thus strongly consider leveraging Global Cost Optimization via Reservations for lift & shift workloads. Restricting the number of possible server locations and virtual machine flavor types (see Resource Configuration Policies) helps consolidating workloads so that you can purchase larger reservations. From our experience this can provide huge savings between 25-65% quickly and with minimal effort.

Currently no tool implementations documented. Contributions welcome!

Depends On