28 Jun Make Your Mainframe A Digital Business Resource
Leaving your mainframe-based applications and data out of your digital business strategy is a bad idea. Unfortunately, buttressed by overhyped notions like bimodal IT, that’s exactly what many enterprises are doing. However, circumstances are about to force every enterprise to make a “migrate or modernize” decision about their mainframe-based applications and data. In most cases, modernize is the better choice. Here’s why.
The Issue: Mainframe Apps And Data Are Really Valuable
A key goal of any digital business transformation is to find ways to grow and leverage data (and applications, which, technically speaking, also are data) that improve customer engagement, lower costs and differentiate the business. Data is a unique asset in that data can be exploited over and over again, and in different forms, without diminishing its value, now or in the future. Creating new options for increasing business value with data is a central objective for a digital business (and should be the objective for today’s CIO).
Are the operational applications and data that typically run on mainframes in large enterprises valuable? Absolutely! Generally, the systems of record that are the basis for monitoring and reporting business performance are assets that any business wants to apply to new digital opportunities, like new products and services (e.g., finance and insurance) or yield optimization (e.g., transportation). Walling them off greatly diminishes digital options and puts your business at a disadvantage to competitors that do a better job of integration.
The Problem: Your Mainframe Staffers Are Retiring
The old mainframe technology guard that helped usher in the digital age are getting on in years and exiting the workforce. To sustain the high-value transactional applications (HVTA) at the core of operations, businesses need younger application delivery professionals to manage mainframe-based applications. But the “digital donut” strategy that explicitly holed-out your mainframe application and data portfolio from your cloud strategy, as suggested by notions like bimodal IT, ensured separation in tooling and practices that make it increasingly challenging to attract and excite new generations of technology professionals to work with core operational applications and data.
That leaves you with two near-term options: (1) substitute mainframe applications and data with other platforms and (2) modernize your mainframe application and data platform. Which is better? That depends on a myriad of circumstances, but in general platform migrations are worse and modernizations are better.
The Superior Solution: Bring Cloud Tools And Practices To Your Mainframe
The first thing to note is this: Hardware should not be the focus of any effort to modernize mainframe applications and data. Of course, hardware is not irrelevant, but in many respects IBM’s “IBM Z” machines provide differentiating capabilities that could enhance digital business options. For example, IBM’s Passport technology builds upon the unique end-to-end (i.e., across any device that touches the transaction) data encryption of IBM Z systems to make it vastly easier to share data across systems of record and data analytics systems without compromising data security and privacy.
To modernize mainframe applications and data, the real goal must be to adopt and master cloud-native tooling and practices to ensure that developers and operators can use common methods and management to speedily exploit new digital opportunities – across all cloud and on-premises platforms. In other words, you need to pull a cloud-native IT operating model over your mainframe applications and data.
What does that mean? In simplest terms, it means bringing popular open source tools for DevOps to the mainframe, extending those tools for use with mainframe-oriented languages and automation services and making a serious commitment to reskill application delivery and operations professionals to fill in the digital donut in cloud operations.
The good news is that IBM is going all-in on cloud, fully committing to end differences between how you deliver applications and services in the cloud and how you’ll deliver application and services on IBM platforms (including IBM’s own cloud offering). IBM’s most recent step on this journey is the introduction of “IBM Wazi for Red Hat CodeReady Workspaces,” which can be an add-on for IBM’s “CloudPak for Applications” (yeah, that’s a mouthful, but it’s IBM), but also can be used standalone without the need to buy the cloudPak for applications. This bundled set of capabilities:
- Elevates the primacy of open source DevOps tooling on the mainframe. Perhaps it took the acquisition of Red Hat, but the “Wazi” DevOps pipeline is coherent and consistent with Red Hat’s leading CI/CD toolsets. The framework will be immediately familiar to any application delivery professional raised on tools like Git, Ansible and Jenkins.
- Integrates COBOL and PL/1 language tools and IBM Z automation. A containerized z/OS sandbox running on RedHat Openshift on x86 Linux brings DevOps tooling and practices to traditional mainframe languages. While Cobol, Pl/1 and assembler programmers using traditional 3270 tools will still face learning curves, the end result will be a common DevOps experience for all application development pros in (or consulting to) your shop. Moreover, not only is IBM introducing new analyze and build tools for using old languages with new DevOps pipeline tricks, Wazi also introduces a separate but complementary new offering, “IBM Wazi Virtual Test Platform,” that allows developers to fully integrate application integration and transaction-level testing into a standard enterprise wide DevOps pipeline, an essential capability for early-and-often (i.e., “shifting left”) hybrid cloud testing without breaking other systems – or the bank.
- Alleviates training and change fears. Old dogs certainly can learn new tricks, but sometimes choose not to. IBM’s “DevOps Acceleration Program” (DAP) and “IBM Z Academic Initiative” DevOps reduces assessment, training and deployment barriers for…