Mission critical systems in government, healthcare, financial, and other industries are trapped on mainframe systems waiting to take advantage of the cost savings, enhanced security, and improved resiliency of the cloud. Hindering these modernization efforts are the systems written in older languages such as COBOL. In a new case study featured on the AWS Public Sector blog, our ReDuX engineers describe how they used our AI toolkit to accelerate transformation.

The Mainframe Modernization Challenge

We are an Amazon Web Services Advanced Tier Services Partner serving United States federal government agencies. One of our agency customers needed to rearchitect their mainframe applications to modern, cloud native services. This was a complex legacy system with tightly coupled applications and limited documentation. 

Moreover, it was a mission critical system and thus needed to maintain availability throughout migration and modernization. Therefore, the team must take an incremental approach, testing and retesting the system in the new environment. Our team found we could accelerate the process from start to finish with AI. 

ReDuX and Amazon Bedrock Accelerate Transformation 

ReDuX is a toolkit consisting of playbooks, mechanisms, and frameworks designed to expedite the modernization of mainframe applications. It incrementally refactors and digitally transforms user experiences, merging valuable aspects of mainframe business-critical applications with new automation needs and design.

Amazon Bedrock powers ReDuX. Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models from leading AI companies via a single API, along with a broad set of capabilities needed to build generative AI applications with security, privacy, and responsible AI. 

The ReDuX AI toolkit analyzed the COBOL source code, providing the team’s business analysts with insights into the system’s domain while generating accurate and complete insights based on that analysis. An AI-enhanced knowledge graph generates a map of the system’s business rules. At this assessment stage, the team could work with a bot to chat with the code, pulling out deeper insights.

Meanwhile, software engineers leveraged code generation capabilities to accelerate their process. This application copilot helps them transform the COBOL code into production ready modern code equivalents. As a result, we found that the developers supported by our AI application copilot had an improved developer experience while boosting their overall productivity.

The complete case study lends insights into these efficiencies, strategies for optimizing this type of AI toolkit for cost efficiency, and further technical details on the process. For this deeper drive from our technical team, read Karsun Solutions builds modernization platform using Amazon Bedrock on the AWS Public Sector Blog. Contact us to connect with our team and start your mainframe modernization journey.