Cloud engineering: The benefits of using a cloud migration factory

Evolving beyond lift-and-shift is critical

The ability to evolve from legacy systems to the cloud can increasingly define an organization’s success. As artificial intelligence (AI), advanced analytics and more advanced digital capabilities take shape, business leaders recognize that the cloud is at the center of agility, flexibility and productivity. In an ideal scenario, it also unleashes innovation and full-scale disruption. 

But getting to return on investment (ROI) can be extraordinarily challenging. PwC’s 2023 Cloud Business Survey found that while 78% of companies have parts of their business already in the cloud, only 10% fully realize the benefits of being a cloud-powered business.

Why the disconnect? Too often, business and IT leaders think about a cloud migration in limited terms. Ultimately, a business's goal is to get legacy assets into the cloud as quickly as possible with minimal modifications — a common technique that’s referred to as “lift and shift” — taking advantage of Infrastructure as a Service (IaaS) functionality from the cloud service providers. This involves moving applications to the cloud without making changes to underlying code or infrastructure. This approach can make it nearly impossible to take advantage of cloud features like automatic scaling of computers and storage.

Frequently, this approach can leave an organization buried in manual, disconnected tasks rather than enabling a strategic operating model. In addition, an organization can wind up with poorly performing applications, inadequate security, people-heavy operations and duplicate IT infrastructure. Lift-and-shift can also lead to squandered money and resources — as well as an inability to support the needs of a digital business, including advanced analytics and AI.

Current approach: why lift-and-shift can limit cloud capabilities

Because lift-and-shift fails to account for crucial application details such as system configurations, itemized dependencies, user permissions and mismatches related to both software and hardware, it cannot take advantage of cloud features such as automatic scaling, dynamic storage and easy-to-use containers, microservices and application programming interfaces (APIs). 

Yet, for many organizations, lift-and-shift can be attractive because it is a fast and inexpensive way to get to the cloud. In a lift-and-shift migration, IT teams can simply copy infrastructure, code and data and drop it into the cloud. Of course, there are times when an individual lift-and-shift makes sense — and helps save dollars — but this approach can compound problems with technical debt such as outdated operating systems, ineffective databases, inconsistent application-to-server mappings and hard-coded security protections.

Lift-and-shift migrations can also be attractive to procurement organizations because they are straightforward to model and price; the industry standard pricing model for an IaaS migration is some number of dollars per server. This pricing mechanism tends to compound an IT department’s issues with technical debt because organizations are encouraged to take the fastest migration approach, and the consequence is higher ongoing cloud operations costs and a higher overall total cost of ownership. 

All of this can lead to confusion about who owns or manages assets, how long a process takes, what costs it incurs, and what impact it has on the overall cloud and IT environment.

These problems can multiply when an organization attempts to adopt various digital technologies — including artificial intelligence, robotics, extended reality, Internet of Things (IoT) and agile web services that require interfaces with modern cloud technologies that are built upon containers and microservices. After an IaaS migration, DevOps processes are no more effective than they were on premise. Software developers looking to adopt an agile, rapid-fire coding framework encounter speed bumps and often full-scale roadblocks.

The challenges of managing a lift-and-shift cloud environment can extend to areas such as monitoring and logging, tracking cloud spend and implementing cloud-based cybersecurity controls. And it can undermine highly interconnected supply chains that require APIs. In the end, an enterprise may face disappointing results — and even technical debt and unforeseen costs that undercut the reason for migrating to the cloud.

A better way: a factory-line approach

Migrating to cloud is more than just moving IT workloads; it's about taking advantage of cloud to help drive business outcomes that are impossible in old IT models. True modernization requires a standardized cloud architecture and consistent design patterns. PwC calls this approach a cloud-powered factory line. It’s characterized by five core stages, each made up of specific methods and solutions that can lead to a better cloud migration.

A factory-line approach greatly reduces manual tasks that require specific and discrete solutions and controls. It does away with silos and the need to manage migrations on a server-by-server basis. Instead, it helps interconnect key processes and stakeholders and standardizes cloud design, thus leading to greater predictability and highly dependable outcomes.

In a real-life example, this approach helped a company that needed to migrate more than 50,000 unmanaged cloud resources to a new platform that supported standardized services, controls and automation.

After analyzing the overall IT framework, PwC began moving largely unmanaged cloud resources to the cloud. The migration-centered factory line tapped the experience of specialized teams for application diagnostics and design, code refactoring, integration with CI/CD pipelines, and finally, data cutover and hypercare support. After the migration, the client was able to take over management of the applications in a more effective operating model.

The final framework supported cloud governance automation and an ultrahigh level of application standardization.

What’s next?

Clouds continue to evolve — and digital technology continues to advance. For example, generative AI can have a major impact on organizations and the public. Adapting applications for these solutions requires that they operate on a highly agile and flexible cloud framework.

To stay at the leading edge of digital innovation and take advantage of emerging technology, it’s critical to build and maintain an advanced cloud architecture. Once that architecture is in place, a factory-line approach can position enterprises to leverage the ongoing investments cloud service providers are making to deploy new cloud functionality. This approach allows applications to operate in the cloud with agility, flexibility and scalability — while lowering overall costs. What’s more, after migration, cloud-powered applications are often better equipped to accommodate whatever changes take place inside or outside an enterprise.