Starting your mainframe modernization journey with Azure

The mainframe's strengths include the ability to process high volumes of transactions, support complex and highly regulated workloads, and ensure security and reliability at scale. By embracing and unlocking the mainframe's potential as a source of reliable, secure and relevant data and information, organizations can unlock new growth opportunities and maintain a competitive edge.

Integrated mainframe and Azure development teams are an essential part of this outcome—combined teams innovate faster and make better and more effective use of resources. The use of AI developer assistance tools, common CI/CD processes and automated testing can also help to streamline the development process.

Also, these tools can help increase security and reduce errors, allowing developers to focus on outcomes and to release high-quality solutions that bring the best of both platforms to market.

Nowadays, organizations focus on business acceleration, productivity, efficiency and security. They adopt a “fit-for-purpose” strategy that places workloads on the best and most appropriate platform, whether it’s mainframe, public cloud, on premises or at the edge. Using this approach, integration rather than migration is now a key priority.

Given the likely impact on established cultures and working practices integration implies, it’s critically important that leaders are able to set out a clear vision of what integration really means. It is essential to highlight the importance of enhanced agility to meet customer demand and the need to build on the strengths of each platform to deliver advanced features and capabilities. Another important aspect is the potential to accelerate innovation through effective collaboration.

·      Enhanced agility and responsiveness: Traditional development cycles associated with mainframes can be lengthy - adopting agile practices, mainframe teams can significantly reduce development cycles, allowing for more frequent updates and faster delivery of new features.

·      Synergy of strengths: Mainframes continue to process and secure transaction data, while Azure's AI capabilities can be leveraged to analyze data, predict outcomes and automate processes.

·      Innovation through collaboration: Agile practices such as continuous integration and continuous deployment (CI/CD), automated testing, iterative development and teams that use common tools and repositories, such as GitHub, can improve quality and efficiency while facilitating knowledge transfer.

Mainframe and Azure development teams often have deeply ingrained practices and mindsets. The challenge lies in convincing stakeholders at all levels of the advantages of consistent processes, assets and automation in the context of their roles and responsibilities. The elimination of silos and barriers removes frustrations, enabling teams to collaborate on higher value outcomes for the business.

Changing established working practices and adopting new tools is not straightforward. It is important to stay focused on the opportunities for improvement and to be able to clearly set out the reasons why this process is necessary.

The vision must be communicated consistently and persuasively, not only through formal communication channels such as meetings and presentations but also informal channels such as one-on-one conversations and team discussions. Leaders must be visible and actively engaged in communicating the vision, addressing concerns and actively providing support.

It’s helpful to start with an assessment, reviewing existing commitments, ongoing and planned projects, existing challenges and issues and the health and culture of development teams. Critical evaluation guides informed decisions about which tools, technologies and processes can be used. Also, this approach can tell whether current circumstances can either constrain or limit the set of options and choices available.

·      Moving quickly to adopt modern programming languages and toolchains such as Java^TM, Python and Ansible might be necessary to stay competitive. This approach can also address resource challenges and better integrate with teams working on applications developed for Azure.

·      The combination of a long backlog of development work and a shortage of experienced COBOL developers might suggest that the introduction of AI-enabled tools such as IBM® watsonx Code Assistant® for Z (WCA4Z) needs to be prioritized.

·      The need for mainframe and Azure development teams to work closely on a strategic initiative can mean that both development teams need to embrace and adopt agile working practices. This process includes attending regular meetings, developing shared milestones and using continuous integration and delivery.

·      A renewed focus on security, compliance and secure code suggests that introducing DevSecOps and comprehensive security monitoring needs to be a priority.

Each organization and team has its own set of requirements, challenges and priorities. Refer here for some of the options available to z/OS® teams interested in moving to DevOps, and wanting to use modern integrated development environments (IDEs), version control, automation and more.

There are clearly numerous different options and approaches to choose from. One of them includes new technology solutions (like AI assistants for example) that can be used to speed up code generation and review and assess code for security vulnerabilities. Another example can be the new and potentially different working practices and approaches to development. It’s important to understand and select the options that best meet the needs of the team and its objectives.

Effective planning and preparation are crucial for the success of any large and complex change initiative. Integrating mainframe and Azure into a strategic platform requires careful consideration. This practice is especially important when such integration entails aligning development teams and key milestones as well.

In prioritizing the most important tasks and ensuring continuous improvement with quick and tangible evidence of progress, teams can achieve a smooth transition that minimizes disruption and maximizes the benefits over time.

Common challenges in shifting to agile practices include resistance to change due to concerns about new methodologies disrupting established workflows. It is essential to plan for and discuss some of the cultural changes required to successfully implement agile practices and be able to address the issues and challenges that come up. Training and education must be included in the transition plan.

Over on GitHub, you can read about some of the planning considerations for z/OS teams interested in moving to DevOps, with special attention given to establishing clear goals and milestones. The process of moving to new working practices is likely to be highly disruptive in the short and medium term. Leaders need to understand and to plan for how to keep current projects moving and maintain existing commitments while this transition is happening.

The sequence to deliver successful integration starts with the aim of increasing individual productivity. As the process advances, it builds toward fully integrated, agile development teams. This method involves AI, agile working practices and new development approaches to create solutions that bind Azure and mainframe into a strategic platform for competitive advantage.

Modern tools and AI are revolutionizing how individual mainframe developers work, significantly boosting productivity and accelerating modernization efforts.

A contemporary and familiar developer experience for today’s professionals that use VSCode with extensions such as IBM’s Z Open Editor brings all the expected accelerators. This practice also enables developers to work with COBOL and other Z-centric languages.

With solutions such as IBM watsonx Code Assistant for Z (WCA4Z), developers can visualize complex COBOL codebases and their dependencies, making it easier to identify and isolate key modules. Generative AI can then be used to convert legacy COBOL into object-oriented Java, enabling developers to extend functionality and prepare applications for integration with modern platforms.

AI-powered assistance in the form of GitHub Copilot can transform the understanding, learning and efficiency for developers that use Java, Go or Python (all now available on z/OS).

As individual productivity improves, the next step is to bring mainframe developers together as a cohesive, agile team. Adopting agile methodologies represents a cultural shift from traditional, siloed development practices to a more collaborative, iterative approach. Teams begin working in short sprints, prioritizing tasks based on business impact and delivering incremental updates that can be quickly tested and refined.

Technology solutions such as Azure Boards offer a platform for agile planning and collaboration. Teams can manage user stories, track progress by using Kanban boards, and potentially link work items to code commits and pull requests. This integration aims to enhance traceability and transparency throughout the development lifecycle and aligns mainframe development with agile, cloud-native practices.

The large and monolithic nature of enterprise applications can often slow down development, both at the organizational level and at the individual developer level. Having started to adopt agile working practices as mentioned before, organizations often start to think about moving and migrating source control to Git repositories (GitHub being one example). This exercise serves as an opportunity to examine how they can modernize, breaking down monolithic applications into smaller components and function-specific code repositories.

Traditional z/OS development tools and processes have historically made it challenging for z/OS applications to be integrated into modern CI/CD pipelines. Now, it’s possible to create toolchains that incorporate z/OS applications into the same CI/CD pipelines used by teams developing nonmainframe applications.

This integration allows z/OS teams to use contemporary, industry-standard, open source technologies for integrated development environments (IDEs), version control, automation and other purposes. This approach proves to be a logical next step when source control has been moved to Git.

For more information about this subject, see Git branching model for mainframe development. Details of the technical implementation and how it is achieved by using IBM Dependency Based Build (DBB) and zAppBuild, are discussed in “Implementing a pipeline for the branching model”.

Bringing mainframe and Azure development teams together is the final and most strategic step in the transformation journey. The use of common tools can help remove many of the barriers to communication, break down silos and transform development efforts.

Such tools include Git and GitHub for source control, integrated developer environments (IDE) like Visual Studio Code, and AI development aides like Watson Code Assist for Z and GitHub Copilot. CI/CD pipelines like Azure DevOps are another good example of useful tools that can help the transformation. Also, the adoption of agile working practices is a necessary step in creating integrated teams. Through these efforts, the day in the life of a developer for individuals working on mainframe applications becomes similar to professionals working in Azure.

Read for more details on how mainframe application teams can implement a developer experience, CI/CD pipeline and DevOps practices that allow them to integrate with teams working on applications developed for Azure.

An integrated mainframe and cloud development team should see improved collaboration, faster development cycles and enhanced code quality. The use of AI tools such as code generation and automated testing can streamline the development process. Also, these tools can reduce errors and foster innovation, ultimately delivering competitive advantage and high-quality solutions.

Measuring success is essential to ensure that the changes are delivering the wanted outcomes and benefits. Typical metrics formed around DORA (DevOps, research and assessment), SPACE (satisfaction, performance, activity, communication and efficiency) and business-specific requirements such as feature cycle time and test coverage are also important. These metrics can be used to measure the effectiveness of new working practices, technology and development tools. Conducting retrospectives at various levels can also be helpful to gather more detailed qualitative feedback and insights from the different teams involved.

• Performance: Evaluate whether teams are seeing better performance by adopting new standards. This process can include measuring the speed and efficiency of development processes, the quality of the code produced, and the overall productivity of the team.

• Activity: Assess whether team members are able to contribute more or faster by adopting new standards. This approach can involve tracking the number of tasks completed, the time taken to complete tasks, and the level of collaboration among team members and across different teams.

• Communication and collaboration: Measure the effectiveness of communication and collaboration among team members. This method can include evaluating how well team members are building on existing standards collaboratively and how effectively they are working together to achieve a common goal.

• Deployment frequency: Compare the relative deployment frequencies across teams to determine how often new code is being released to production. Higher deployment frequencies can indicate a more efficient and agile development process.

• Lead time for changes: Identify where changes stall in the software development lifecycle (SDLC) and measure the time taken to implement changes. Shorter lead times can indicate a more responsive and agile development process.

• Change failure rate: Compare deployment failure rates across teams and identify areas for improvement. Lower failure rates can indicate higher code quality and more effective testing and deployment processes.

Integrating mainframe and Azure development to achieve optimal results requires focus, clear outcomes and methodical, gradual changes rather than drastic ones. A combination of changes to roles, responsibilities, processes, the introduction of (new) technologies and of helping different development teams has come together to deliver real tangible outcomes. It’s time to reflect on the progress made, evaluate next steps and determine how far you can go and how much more you can do. More details can be read here.

IBM’s collaboration with Microsoft Azure provides a robust framework for mainframe application modernization, leveraging Azure’s scalable cloud infrastructure and analytics alongside the mainframe’s reliability and security. This partnership enables businesses to seamlessly switch to a hybrid cloud model, combining IBM’s AI-driven solutions with Azure’s services to enhance agility, reduce operational costs and accelerate digital transformation.

To see what this looks like for your organization and to unlock the full potential of mainframe and Azure integration, schedule a conversation with our experts. Also, you can explore how IBM and Microsoft can help you achieve your goals.