Mainframe Modernization in the Financial Services Industry

The Changing Role of the Mainframe

The continued prevalence and reliance of mainframe in financial services organizations comes down to critical characteristics of that platform that include resilience, speed, security, and scale.  But while the mainframe continues to be a critical resource in the institution's infrastructure, it doesn't mean that the platform is standing still in its evolution or role it plays in the institution.  The technological advances that have happened over the last decade, including AI and advanced digital channels, are driving a demand to ensure that business applications are delivered on fit-for-purpose platforms that allow the institution to become or remain agile and respond to market conditions as quickly as possible.

Cloud, for example, has become a viable alternative for many workloads as it, too, brings a number of beneficial characteristics to the institution’s infrastructure, including faster scaling, usage-based pricing, a constantly growing software ecosystem, and shifting spend from capital to operational budgets.  Surprisingly, although most organizations list security as a concern when moving to cloud,  according to IDC surveys financial services organizations report improved IT security for workloads that have migrated to cloud.  By no means does this mean that every application or workload can be, or even should be, migrated from the mainframe to a cloud.  But the combination of business drivers like AI and the benefits of modern development paradigms and deployment models leads inexorably to a re-evaluation of the role of the mainframe, and to what extent the mainframe should evolve to continue to support the needs of the institution.

The reasons an institution would consider modernization include, but are not limited to:

• A new technology like AI can drive the institution to update the mainframe or deploy applications on cloud to support the new technology.
• New workloads are vetted based on the sensitivity of the data they produce or consumer.  This may drive some modernization of applications to support the new functionality, which may, in turn, lead to either moving the application to cloud, or updating the mainframe to deploy the workload there.
• The cost of mainframe operations is often cited, justifiably, as a reason to modernize workloads and move them to non-mainframe environments.
• Talent for mainframe development and support is also a hurdle for every financial institution that operates the platform.  In particular, the need for COBOL development to maintain and, in some cases, create new applications is stretching the institution's ability to find and retain staff that can work on COBOL/mainframe environment.
• Ultimately, moving to cloud-native environments can enable faster development cycles through modern DevOps practices, enabling the lines of business to compete more effectively in fast-moving and sometimes disruptive markets

Agility and Consistent Methodology Drives Mainframe Modernization Success

The foundational principle of every modernization initiative in financial services should be to serve the needs of the business.  This is true even for technologies like AI that have captured the attention of every executive at the institution.  At an even deeper layer, “agility” should be the battle cry of technology executives that have to carry out the modernization of any platform or workload at the institution.  At a minimum, this means using APIs and microservices to transform often monolithic workloads into sets of open, reusable, and scalable components that span the enterprise’s business architecture and enable fast deployment of new functionality with less development.

The steps that any institution takes to modernize its applications will vary widely, but typically consist of these 5 general steps:

1. Document the application.  This may seem like a trivial step, but the financial services industry is too often challenged by “legacy” applications that are decades old, with little original documentation, and worse, so much customization over the years that any documentation is woefully inadequate to use as a basis of modernization.
2. Discover (or uncover) business functionality.  A critical step in modernizing an application is to understand what it does.  This sounds obvious, but often (and more so in older workloads) institutions will find that the application actually does more than it “should” be doing, or is not actually providing the intended functionality in the most efficient way possible.  In some cases, once the functional audit has taken place, no functional changes are needed.
3. Refactor the workload. Having documented the workload, and understood the functionality buried in the code, the work of refactoring the application can begin.  This stage will require tools that can take advantage of the outputs of the first steps so that developers can leverage the documentation and functional maps of the workloads to help in the development of new code.
4. Deploy on the appropriate platform.  At some point, given the needs of the institution and the characteristics of the infrastructure, the institution will decide where the refactored application should run.  The beauty of using APIs and microservices is that the institution will have the agility to initially migrate the workload to what is considered the appropriate platform, but then redeploy if necessary at a later date.
5. Continuously monitor the modernized workloads in production to ensure that the performance meets all service level agreements and that the deployment environment is meeting performance, resilience, scalability, and cost expectations.

What next?

The challenge of modernization is in the enormous amount of code that needs updating. The steps outlined above will require just as much focus on skilled resources as the initial condition of operating legacy workloads do.  These workloads are often massive in scale, with millions of lines of code.  The ability to modernize them is enhanced by the use of tenured partners, knowledgeable in the financial services industry, who can bring modern tools to bear.

In this way, the cost of resources needed to modernize should be seen as an investment in the future agility of the business.  Modernization is tied directly to business agility, particularly during times of disruption, by enabling the organization to respond to necessary changes in customer expectations, new products demanded by the market, changes in the regulatory environment, and competitive differentiation from faster innovation for first-movers.