Using Collaboration (not Migration) to Modernize Your Mainframe Applications with the Cloud

A look at the various mainframe/cloud collaboration scenarios to see how best to provide value, save money and push digital transformation.

This is part two in a five-part series on mainframe modernization.

When you hear the phrase “modernize your mainframe applications with the cloud,” the first thing to come to mind is likely going to be migration. Not so fast. When you cut through the misconceptions, migrating off the mainframe is rarely the best approach. Today’s mainframes are blisteringly fast, remarkably scalable and so unbelievably reliable that many mainframes in operation today have been running for decades with absolutely no downtime whatsoever. That’s right. Not five 9’s. Not six 9’s. All the 9’s—we’re talking 100% uptime. It’s no wonder two-thirds of the Fortune 100 relies upon its big iron for its mission-critical transactions. Banks, airlines, insurance companies, retailers—the list goes on. The global economy depends on mainframes, every minute of every day. But (and you knew there was a but), these same enterprises are in the midst of digital transformations. New economic and customer pressures are forcing them to innovate and rethink how they leverage technology to deliver customer value. Such transformations invariably include the cloud, and when it does, modernizing the mainframe soon enters the conversation. While AWS, Microsoft Azure and other major cloud providers would love for all the enterprise workloads to run on their clouds, they realize that most mainframe customers would be better served by a mainframe/cloud collaboration strategy. IBM—the sole remaining mainframe vendor and a cloud provider in its own right—also champions bringing mainframes and clouds together to meet the modern digital needs of enterprises that have been depending on the mainframe platform for so many years. Still think mainframe application modernization necessarily means migration? Let’s take a look at the various mainframe/cloud collaboration scenarios to see how best to provide value, save money, and most of all, achieve the customer-centric goals of digital transformation.

Placing mainframe modernization into context

For some enterprises with mainframes, modernization may not be a priority. Following the adage “if it ain’t broke, don’t fix it,” some mainframe applications continue to chug away, providing as much value today as they did when they were new. Leaving such apps alone may be the best business decision, but might also still slow down the organization’s ability to innovate to meet changing customer needs. It’s important to weigh the pros and cons of making any change vs. maintaining the status quo. In other situations, mainframe application modernization initiatives involve updating or replacing older applications on the mainframe with new or reworked mainframe applications. More likely than not, however, mainframe modernization requires some combination of mainframe and cloud-based capabilities working together. Refactoring mainframe applications to run in the cloud is a sometimes considered modernization strategy. Convert COBOL to Java, for example, either to run as-is in the cloud or as part of a cloud-native re-architecture initiative. Reducing mainframe MIPS costs is often a business motivation for shifting some or all applications to the cloud. “Customers want to do more with less,” explains Steven Steuart, AWS WW GTM Mainframe. “Our customers can transform with or to AWS (for example, to reduce MIPS consumption), process on mainframe, and consume on AWS.”

Separating application and data concerns

Enterprises use mainframes to both run applications and store data. The modernization considerations for each purpose are often different. Approaching the question of mainframe vs. cloud as a question of which tool is right for which job is central to cost-effective modernization decisions. Leaving core transaction processing on the mainframe, for example, is a straightforward, low-risk decision, while running workloads like analytics and customer experience apps on the cloud will reduce costs and deliver the full power of cloud-based services. In fact, in many situations, the question is less about where applications are running and more about the data. Supporting mobile apps with mainframe data, for example, can be both expensive and slow due to mainframe processing costs, data transport costs and network latency issues. Selectively replicating mainframe data to the cloud to support read-only access can solve such problems but is only appropriate where real-time access to the data is less important. Inferencing in real-time closest to where the data resides is a competitive advantage the modern IBM z16 mainframe provides. In other situations, the goal is to expand access to mainframe data to cloud-based applications and services. Organizations are thus able to leverage the value of data on the mainframe across their IT landscape.

Mainframe development in the cloud

As the boomer generation of mainframe developers retires, mainframe-based organizations must energize a new workforce. Such professionals, however, don’t want to sit in front of green-screen terminals. They want to work with modern development tools in a modern, cloud-based environment. Enter IBM Wazi Developer for Workspaces. Wazi is a development environment with a browser-based IDE that developers and testers can use to build, test and run mainframe applications from anywhere. AWS, for one, is all-in with Wazi. “Wazi from IBM is available today on the AWS Marketplace as part of IBM’s Z and Cloud Modernization Stack offering,” AWS’s Steuart points out. Wazi addresses the generational mainframe skills problem by delivering a modern experience for developers working with z/OS mainframe apps in the cloud. In addition, IBM has rolled out IBM Wazi as a Service on the IBM Cloud, bringing z/OS development and test as-a-service to the cloud for the first time. “Developers now have easy access to modern development tools and innovative cloud services,” said Andy Bradfield, vice president of IBM Z Hybrid Cloud. “And with hybrid clouds, they can keep their applications wherever they need to be—in the cloud, on-premises and at the edge.”

The Intellyx take

In addition to IBM and the major cloud providers, there’s an entire ecosystem of both young and mature vendors providing tools and platforms to help enterprises modernize their mainframe applications by working with the cloud. Some vendors focus on DevOps tools. Others on mainframe-to-cloud integration. DataOps and other modern data tooling are also available from many ISVs. As a result, the number of mainframe-based enterprises who seek to migrate off of the venerable platform is actually going down as it becomes increasingly clear that mainframes are an integral and essential part of the modern cloud-based world. To learn more, see the other posts in this series: