|
In Might, 2025, we launched AWS Rework for mainframe, the primary agentic AI service for modernizing mainframe workloads at scale. The AI-powered mainframe agent accelerates mainframe modernization by automating complicated, resource-intensive duties throughout each section of modernization—from preliminary evaluation to ultimate deployment. You possibly can streamline the migration of legacy mainframe functions, together with COBOL, CICS, DB2, and VSAM to fashionable cloud environments—reducing modernization timelines from years to months.
At the moment, we’re saying enhanced capabilities in AWS Rework for mainframe that embrace AI-powered evaluation options, assist for the Reimagine modernization sample, and testing automation. These enhancements remedy two essential challenges in mainframe modernization: the necessity to utterly rework functions reasonably than merely transfer them to the cloud, and the in depth time and experience required for testing.
- Reimagining mainframe modernization – This can be a new AI-driven method that utterly reimagines the shopper’s software structure utilizing fashionable patterns or shifting from batch course of to real-time capabilities. By combining the improved enterprise logic extraction with new knowledge lineage evaluation and automatic knowledge dictionary technology from the legacy supply code via AWS Rework, clients rework monolithic mainframe functions written in languages like COBOL into extra fashionable architectural types, like microservices.
- Automated testing – Prospects can use new automated check plan technology, check knowledge assortment scripts, and check case automation scripts. AWS Rework for mainframe additionally gives practical testing instruments for knowledge migration, outcomes validation, and terminal connectivity. These AI-powered capabilities work collectively to speed up testing timelines and enhance accuracy via automation.
Let’s be taught extra about reimagining mainframe modernization and automatic testing capabilities.
Find out how to reimagine mainframe modernization
We acknowledge that mainframe modernization shouldn’t be a one-size-fits-all proposition. Whereas tactical approaches deal with augmentation and sustaining present programs, strategic modernization provides distinct paths: Replatform, Refactor, Exchange, or the brand new Reimagine.
Within the Reimagine sample, AWS Rework AI-powered evaluation combines mainframe system evaluation with organizational information to create detailed enterprise and technical documentation and structure suggestions. This helps protect essential enterprise logic whereas enabling fashionable cloud-native capabilities.
AWS Rework gives new superior knowledge evaluation capabilities which might be important for profitable mainframe modernization, together with knowledge lineage evaluation and automatic knowledge dictionary technology. These options work collectively to outline the construction and which means to accompany the utilization and relationships of mainframe knowledge. Prospects achieve full visibility into their knowledge panorama, enabling knowledgeable decision-making for modernization. Their technical groups can confidently redesign knowledge architectures whereas preserving essential enterprise logic and relationships.
The Reimagining technique follows the precept of human within the loop validation, which signifies that AI-generated software specs and code resembling AWS Rework and Kiro are constantly validated by area consultants. This collaborative method between AI capabilities and human judgment considerably reduces transformation threat whereas sustaining the pace benefits of AI-powered modernization.
The pathway has a three-phase methodology to remodel legacy mainframe functions into cloud-native microservices:
- Reverse engineering to extract enterprise logic and guidelines from present COBOL or job management language (JCL) code utilizing AWS Rework for mainframe.
- Ahead engineering to generate microservice specification, modernized supply code, infrastructure as code (IaC), and modernized database.
- Deploy and check to deploy the generated microservices to Amazon Net Companies (AWS) utilizing IaC and to check the performance of the modernized software.
Though microservices structure provides important advantages for mainframe modernization, it’s essential to know that it’s not one of the best answer for each situation. The selection of architectural patterns ought to be pushed by the particular necessities and constraints of the system. The secret’s to pick an structure that aligns with each present wants and future aspirations, recognizing that architectural selections can evolve over time as organizations mature their cloud-native capabilities.
The versatile method helps each do-it-yourself and partner-led growth, so you need to use your most well-liked instruments whereas sustaining the integrity of your small business processes. You get the advantages of recent cloud structure whereas preserving many years of enterprise logic and lowering venture threat.
Automated testing in motion
The brand new automated testing characteristic helps IBM z/OS mainframe batch software stack at launch, which helps organizations deal with a wider vary of modernization eventualities whereas sustaining constant processes and tooling.
Listed here are the brand new mainframe capabilities:
- Plan check instances – Create check plans from mainframe code, enterprise logic, and scheduler plans.
- Generate check knowledge assortment scripts – Create JCL scripts for knowledge assortment out of your mainframe to your check plan.
- Generate check automation scripts – Generate execution scripts to automate testing of modernized functions operating within the goal AWS surroundings.
To get began with automated testing, it’s best to arrange a workspace, assign a selected position to every consumer, and invite them to onboard your workspace. To be taught extra, go to Getting began with AWS Rework within the AWS Rework Consumer Information.
Select Create job in your workspace. You possibly can see all forms of supported transformation jobs. For this instance, I choose the Mainframe Modernization job to modernize mainframe functions.
After a brand new job is created, you may kick off modernization for assessments technology. This workflow is sequential and it’s a place so that you can reply the AI agent’s questions, offering the mandatory enter. You possibly can add your collaborators and specify useful resource location the place the codebase or documentation is situated in your Amazon Easy Storage Service (Amazon S3) bucket.
I take advantage of a pattern software for a bank card administration system because the mainframe banking case with the presentation (BMS screens), enterprise logic (COBOL) and knowledge (VSAM/DB2), together with on-line transaction processing and batch jobs.
After ending the steps of analyzing code, extracting enterprise logic, decomposing code, planning migration wave, you may expertise new automated testing capabilities resembling planning check instances, producing check knowledge assortment scripts, and check automation scripts.
The brand new testing workflow creates a check plan in your modernization venture and generates check knowledge assortment scripts. You’ll have three planning steps:
- Configure check plan inputs – You possibly can hyperlink your check plan to your different job recordsdata. The check plan is generated primarily based on analyzing the mainframe software code and might present extra particulars optionally utilizing the extracted enterprise logic, the technical documentation, the decomposition, and utilizing a scheduler plan.
- Outline check plan scope – You possibly can outline the entry level, the particular program the place the appliance’s execution circulate begins. For instance, the JCL for a batch job. Within the check plan, every practical check case is designed to begin the execution from a selected entry level.
- Refine check plan – A check plan is made up of sequential check instances. You possibly can reorder them, add new ones, merge a number of instances, or cut up one into two on the check case element web page. Batch check instances are composed of a sequence of JCLs following the scheduler plan.
Producing check knowledge assortment scripts collects check knowledge from mainframe functions for practical equivalence testing. This step actively generates JCL scripts that can assist you to collect check knowledge from the pattern software’s numerous knowledge sources (resembling VSAM recordsdata or DB2 databases) to be used in testing the modernized software. The step is designed to create automated scripts that may extract check knowledge from VSAM datasets, question DB2 tables for pattern knowledge, gather sequential knowledge units, and generate knowledge assortment workflows. After this step is accomplished, you’ll have complete check knowledge assortment scripts prepared to make use of.
To be taught extra about automated testing, go to Modernization of mainframe functions within the AWS Rework Consumer Information.
Now out there
The brand new capabilities in AWS Rework for mainframe can be found in the present day in all AWS Areas the place AWS Rework for mainframe is obtainable. For Regional availability, go to the AWS Companies by Area. At the moment, we provide our core options—together with evaluation and transformation—without charge to AWS clients. To be taught extra, go to AWS Rework Pricing web page.
Give it a attempt within the AWS Rework console. To be taught extra, go to the AWS Rework for mainframe product web page and ship suggestions to AWS re:Publish for AWS Rework for mainframe or via your traditional AWS Help contacts.
— Channy
