Mainframes: The last frontier of cloud migration

With cloud now occupying center stage, one might think of mainframes as some aging actor who’s stayed too long on stage. But that’s not true at all. In fact, more than 70 percent of Fortune 500 companies still run business-critical applications on mainframes, and many companies and institutions have legacy mainframes at the core of their data centers.1 But what’s their role in the age of cloud? Can cloud and mainframe work together to deliver a great performance?

It’s not necessarily an easy question. On the one hand, in spite of the age of mainframes, they’re still really good at what they’ve always been good for: handling thousands of simultaneous users and hosting many critical enterprise “non-stop” applications. And they hold essential business data. On the other hand, because of mainframes’ downside—including high technical debt resulting in slow application change and high costs—many companies and government agencies are eying mainframes as the “last frontier” of cloud migration and modernization.

With many mainframe applications interconnected through common data, Accenture’s modernization approach is a holistic assessment of mainframe applications that creates a value-based, prioritized list of applications to migrate and modernize. The solutions range from emulation for low-ROI to reimagining via AWS cloud-native services for high-ROI applications.

Why migrate?

For many companies, the cost of maintaining a mainframe ecosystem is outpacing revenue growth. So, cost reduction is a big incentive for making the move to cloud, especially for large enterprises. One recent survey found that almost all respondents stated that their IT costs could drop by up to 50% by implementing cloud apps and architectures.2 But decision-makers should consider several other benefits, well:

• Reduce risk: Over time, many mainframe applications have become overly complex and difficult to change. Maintenance is now difficult, and even small code changes can be risky—slowing the business down and making it brittle in the face of needed change.

• Increase agility and scalability: Flexibility and responsiveness can be improved through reduced development cycles via continuous integration and continuous delivery (CI/CD), and virtually unlimited infrastructure resources consumed on demand.

• Access critical data: Customers get to leverage their mainframe data, which can contain decades of business transactions, and use that data to feed analytics or machine learning initiatives that can deliver competitive advantage.

• Close the skills gap: Modernization can counter the problem many companies are facing as their mainframe experts reach retirement age, and can also attract new talent interested in modernizing core business workloads.

What are the challenges?

Where are organizations typically stumbling with their mainframe migrations? Many challenges have to do with insufficient experience: Poor project phasing, suboptimal choices when it comes to migration processes, and testing phase challenges that require deep end-state platform expertise.

Most organizations are going to need some help. For that reason, AWS has created a new AWS Mainframe Migration category within its AWS Migration Competency to validate AWS Partners with mature practices and a track record of successful mainframe workload migrations and modernizations. (Accenture is a launch partner for the AWS Mainframe Modernization and one of the first consulting firms to earn this validation.) With this competency as a foundation, AWS Partners like Accenture can develop clients’ internal skills and help build the foundations required for executing complex migrations to AWS. Certified integrators have years of experience in reducing friction when migrating legacy applications.3

What’s your pattern?

Every company’s mainframe ecosystem is unique and has evolved differently over the last 20 to 30 years. So, part of the guidance needed on any migration journey is to help determine the “pattern” you should follow to migrate each application. We call this list of patterns “The Six R’s”:

1. Retain for its useful life, remediating any specific pain points and moving development and test environments to a cloud infrastructure to increase agility and reduce costs.
2. Replace with a package (including commercial-off-the-shelf [COTS] or Software as a Service) that offers improved functionality; extract and migrate data to a new system to reduce complexity and costs.
3. Re-host to a less-expensive location without changing the code, to gain cost benefits without the risk that can result from programming language changes.
4. Re-platform to a different platform/operating system (“lift and shift”), without changing the programming language, and thus permit applications to run in the cloud.
5. Re-factor from legacy code to a modern programming language, using (semi) automated tooling to mitigate risk relating to legacy skills, increase agility and reduce costs.
6. Re-imagine the business: Rewrite the application based on newly developed requirements. The resulting application goes beyond current capabilities, allowing for technology modernization as well as the modernization of what were likely outdated business processes.

A proper assessment, working with your IT team, helps determine what variety of patterns is right for you and your applications. The actual migration work then takes place, following an “assess, mobilize, migrate” structure (see figure). Especially helpful to this migration work is the AWS Migration Acceleration Program (MAP), designed in collaboration with Accenture, which helps accelerate modernization planning and execution, offsetting costs incurred during the modernization process.

Success story: AWS mainframe migration

Accenture Federal Services (Accenture) worked with the US Department of Education’s office of Federal Student Aid (FSA) on an AWS mainframe migration. FSA manages Common Origination and Disbursement (COD), a mission-critical suite of applications that processes federal financial aid for those pursuing postsecondary education either in the US or abroad.

To save money, improve business agility and enhance information security, FSA wanted to modernize COD by re-architecting its applications from its mainframe platform to a fully automated, modern technology stack hosted on a FedRAMP authorized cloud service provider, AWS GovCloud (US).

The team used automation, containerization, and a flexible cloud architecture to build COD’s cloud platform, which now includes over 80 applications and supports more than 25 releases per year.

A step at a time

One way to think about mainframe modernization is that, even with a migration, you are still leveraging the value of mainframes—in fact, extending their value through augmentation. Fully modernizing the mainframe ecosystem is a multi-year journey, but important steps can and should be taken now. By modernizing low-risk, low-complexity applications first, you can build the credibility and confidence to establish and maintain momentum through to completion. Accenture’s modernization accelerators and extensive experience with mainframe modernization, combined with AWS technology and financial programs, provide a compelling package to help clients reimagine their mainframe footprint in the cloud.