6 Strategies for Migrating Applications to the Cloud

This post outlines 6 different migration strategies we see customers implement to migrate applications to the cloud.

Formulating a Migration Strategy

Enterprises typically begin to contemplate how to migrate an application during the second phase of the “Migration Process” — Portfolio Discovery and Planning. This is when they determine what’s in their environment, what are the interdependencies, what’s going to be easy to migrate and what’s going to be hard to migrate, and how they’ll migrate each application.

Using this knowledge, organizations can outline a plan (which should be considered subject to change as they progress through their migration and learn) on how they’ll approach migrating each of the applications in their portfolio and in what order.

The complexity of migrating existing applications varies, depending on the architecture and existing licensing arrangements. If I think about the universe of applications to migrate on a spectrum of complexity, I’d put a virtualized, service-oriented architecture on the low-complexity end of the spectrum, and a monolithic mainframe at the high-complexity end of the spectrum.

I suggest starting with something on the low-complexity end of the spectrum for the obvious reason that it will be easier to complete — which will give you some immediate positive reinforcement or “quick wins” as you learn.

6 Application Migration Strategies: “The 6 R’s”

The 6 most common application migration strategies we see are:

1. Rehosting — Otherwise known as “lift-and-shift.”

We find that many early cloud projects gravitate toward net new development using cloud-native capabilities, but in a large legacy migration scenario where the organization is looking to scale its migration quickly to meet a business case, we find that the majority of applications are rehosted. GE Oil & Gas, for instance, found that, even without implementing any cloud optimizations, it could save roughly 30 percent of its costs by rehosting.

Most rehosting can be automated with tools (e.g. AWS VM Import/Export, Racemi), although some customers prefer to do this manually as they learn how to apply their legacy systems to the new cloud platform.

We’ve also found that applications are easier to optimize/re-architect once they’re already running in the cloud. Partly because your organization will have developed better skills to do so, and partly because the hard part — migrating the application, data, and traffic — has already been done.

2. Replatforming — I sometimes call this “lift-tinker-and-shift.”

Here you might make a few cloud (or other) optimizations in order to achieve some tangible benefit, but you aren’t otherwise changing the core architecture of the application. You may be looking to reduce the amount of time you spend managing database instances by migrating to a database-as-a-service platform like Amazon Relational Database Service (Amazon RDS), or migrating your application to a fully managed platform like Amazon Elastic Beanstalk.

A large media company we work with migrated hundreds of web servers it ran on-premises to AWS, and, in the process, it moved from WebLogic (a Java application container that requires an expensive license) to Apache Tomcat, an open-source equivalent. This media company saved millions in licensing costs on top of the savings and agility it gained by migrating to AWS.

3. Repurchasing — Moving to a different product.

I most commonly see repurchasing as a move to a SaaS platform. Moving a CRM to Salesforce.com, an HR system to Workday, a CMS to Drupal, and so on.

4. Refactoring / Re-architecting — Re-imagining how the application is architected and developed, typically using cloud-native features.

This is typically driven by a strong business need to add features, scale, or performance that would otherwise be difficult to achieve in the application’s existing environment.

Are you looking to migrate from a monolithic architecture to a service-oriented (or server-less) architecture to boost agility or improve business continuity (I’ve heard stories of mainframe fan belts being ordered on e-bay)? This pattern tends to be the most expensive, but, if you have a good product-market fit, it can also be the most beneficial.

5. Retire — Get rid of.

Once you’ve discovered everything in your environment, you might ask each functional area who owns each application. We’ve found that as much as 10% (I’ve seen 20%) of an enterprise IT portfolio is no longer useful, and can simply be turned off. These savings can boost the business case, direct your team’s scarce attention to the things that people use, and lessen the surface area you have to secure.

6. Retain — Usually this means “revisit” or do nothing (for now).

Maybe you’re still riding out some depreciation, aren’t ready to prioritize an application that was recently upgraded, or are otherwise not inclined to migrate some applications. You should only migrate what makes sense for the business; and, as the gravity of your portfolio changes from on-premises to the cloud, you’ll probably have fewer reasons to retain.

Cloud DevOps Admin Guide