For more than 50 years, mainframes have been the powerhouse for most global corporations’ large-scale transaction processing. High availability, reliability, security, and serviceability are fundamental tenets for mainframes longevity with organizations. Mainframes handled different instances of an operating system at the same time and thus pioneered virtualization, which has become the de facto principle of cloud environments. 

With the cloud going mainstream, enterprises dependent on mainframes have two big problems to address. On the one hand, they need to compete with new age, born-in-the-cloud companies that are disrupting age-old enterprises with their customer experience-driven differentiated solution. On the other hand, the legacy mainframe technology does not provide the necessary agility for speed and time to market. Additionally, organizations are facing the challenges of mainframe professionals reaching retirement age. Thus organizations risk losing their subject matter experts’ knowledge. Moreover, with an augmented Gen Z workforce, legacy mainframe toolsets are a deterrent in their adoption.

The worldwide COVID-19 outbreak has further accelerated the need for mainframe customers to shift to cloud technologies as physical boundaries have become irrelevant. The new normal mandates cost optimization of MIPS, 24×7 remote connectivity, and the ability to scale up and down dynamically. 

With the above challenges, mainframe clients need a strategy that makes them more nimble to take advantage of Agile-based software delivery and modern toolsets and eventually move to the cloud.

Pathways for Mainframe Modernization

Organizations have two paths for modernizing their mainframe. A business’ use cases, and the timeline to meet them, should help govern the selection.

• In-place upgrade

As part of an “in-place upgrade” strategy, you can upgrade the codebase to a newer compiler or a brand new mainframe to take advantage of new features or move the dev/test workload to Azure to provide agility via a DevOps toolchain and get speed to market advantage. Additionally, you can replicate transactional mainframe data to the cloud for building/integrating solutions. Moreover, batch and scheduler optimizations can significantly reduce the MIPS cost.

• Exiting the mainframe

When it comes to a strategy for “exiting the mainframe,” you can consider re-hosting the mainframe workloads to virtual machines on Azure or replacing the workloads with a suitable SaaS-based solution. You can also re-engineer the COBOL codebase to Java or .NET, move the mainframe data to a data warehouse in the cloud, or re-architect the application to become cloud-native.

The Value Proposition of Azure for Mainframe Modernization

Azure provides mainframe-like features that customers are looking for without the drawbacks associated with mainframes. Azure’s benefits include:

• High Availability – With local and geo-based replication, you can be “always-on” without worrying about your mainframe going down.
• Flexibility as Choice – Auto scale-up and scale-down based on your consumption without worrying about procuring new mainframe hardware to address spikes.
• Isolated Environments – Run varied workloads built on different technologies isolated through a virtual network and applied policies without worrying about the creation of logical partitions (LPAR) on the mainframe.
• Reduced Cost – By transitioning from a mainframe (CAPEX) to an Azure (OPEX) model, you get more headroom for innovation instead of spending to keep the lights on.

Organizations can benefit significantly by moving the mainframe workloads from their private data center to a shared environment offered by a cloud provider like Azure. More than 90 percent of Fortune 500 companies are already using Azure for critical workloads. With its enterprise-grade security and unlimited ability to scale, mainframe clients now have a strong reason to move to Azure (where it makes sense based on mainframe workload characteristics).

Our Recommendations for Mainframe Modernization Journeys

Although you can pursue different alternatives, we strongly recommend abiding by the following principles to achieve your business goals:

• Have a clear roadmap to your short-term and long-term desired states that aligns with your business priorities.
• There is no one size fits all approach when it comes to mainframe modernization. Make a selection and tailor it to the use cases you are pursuing.
• Subject matter expertise and knowledge are diminishing with aging mainframe practitioners. Invest in creating tool-based documentation for your legacy applications.
• Start small, piloting a specific use case before finalizing your path.
• Think through how you can harvest transactional data sitting on the mainframe to the cloud and create new value-added services.
• Find an opportunity for re-imagining the core business process as a long-term solution to become more nimble.
• Cater to Gen Z in your workforce by adopting modern Dev/Test toolsets and building a continuous delivery pipeline backed by a DevOps toolchain on the mainframe.
• Monetize core mainframe functions to take advantage of the API economy.

Manoj Tiwary

Lead Architect, Cloud Services, Wipro

Manoj is the Azure Ecosystem Practice Lead and, as part of his charter, working on GTM solutions, practitioners enablement, pre-sales, and sales support of Azure. He is an accomplished information technology leader with 21 years of progressive leadership experience, having a proven track record in technology consulting, capability building, new business development, delivery management, people management, and deeper client engagement.