Google’s Gemini AI Will Help Rewrite Mainframe Apps and Move Them to the Cloud

Shortly before the announcement of the new IBM z17 mainframes, Google announced new AI tools based on Gemini models and other technologies for modernizing infrastructure and moving workloads to the Google Cloud.

Google Cloud Mainframe Assessment Tool (MAT) is now available, powered by Gemini AI models. The tool allows you to assess and analyze the overall health of your mainframes, including applications and data. This will enable you to make informed decisions on optimal upgrades. MAT provides deep code analysis, generates clear explanations of how it works, automates the creation of documentation, etc. This allows you to quickly understand mainframe code and stimulate the modernization process.

Google Cloud Mainframe Rewrite, powered by Gemini models, enables modernization of mainframe applications (now in preview). It helps developers rethink and transform mainframe code by rewriting it in modern programming languages ​​like Java and C#. Mainframe Rewrite provides an IDE for modernizing code, testing, and deploying modernized applications to Google Cloud.

Image source: Ant Rozetsky / Unsplash

Finally, to reduce the risks associated with modernization, Google Cloud Dual Run is a tool for deep testing, certification, and assessment of modernized applications. The tool allows you to check the correctness, completeness, and performance of the modernized code during the migration and before the new application replaces the old one. Dual Run compares the data produced by the old and new systems to find differences.

There are also additional tools developed by Google partners. For example, Mechanical Orchard offers a platform for quickly rewriting COBOL applications into modern languages, including Java, Python, etc., without changing the business logic. The solution allows for “step-by-step” rewriting of application fragments using generative AI systems while maintaining functionality and testing the correctness of their operation. The main goal is to create a functional equivalent of legacy solutions for the cloud.

Image source: Google Cloud

In addition, as part of the new Google Mainframe Modernization with Gen AI Accelerator program, the company has brought in Accenture, EPAM, and Thoughtworks to help organizations migrate from mainframes. The first stage involves code analysis using MAT and Gemini. The second stage involves a pilot project to test the functionality of the new code. The third stage involves a full-scale migration to the cloud. The first customers can evaluate the solutions for 4-8 weeks for free (excluding Google Cloud costs).

It’s not too early to write off mainframes. According to a Kyndryl 500 survey of IT industry executives, many organizations are integrating mainframes with public and private cloud platforms and are enhancing their modernization programs by moving some workloads off mainframes or upgrading them.