Navigating Modernization Complexity? Truly Unbiased Mainframe Experts Want to Help

Summary

  • For many, budgets are tight, bringing mainframe costs into focus
  • The way we see it, the “easy” mainframe workload migrations have already taken place. (There may have been budget or timeline problems, but they’re done.) That leaves the harder ones
  • There are a lot of claims out there. Many of today’s executives didn’t grow up on the mainframe – so they don’t have the technical background to know what’s feasible and what’s “aspirational
  • GTSG wants to help. After 35 years, we know the platform cold and can provide an honest look at the proposals you’re getting from mainframe modernizers – or, better still, upstream from proposals, input to help frame your investigation of the available options
  • We’re technology agnostic. Plenty of mainframe workloads have already moved, and more will move in the future. We simply want to help our clients reach a fact-based decision and develop a realistic plan, budget, and timeline while avoiding unnecessary and expensive mistakes along the way.

The Resilient, Evolving Mainframe

The mainframe has been an easy target for a long time, with experts predicting its demise well before the turn of the millennium.[i] Yet mainframes handle 90 percent of all credit card transactions and 68 percent of the world’s production IT workloads, accounting for only 6 percent of IT costs. 71% of Fortune 500 companies use mainframe systems to handle their most critical transactions and processing workloads. Mainframe shops include 44 of the top 50 banks, 23 of the top US 25 retailers, all top 10 insurers worldwide, the Social Security Administration, Internal Revenue Service, and Centers for Medicare and Medicaid Services.

The technology isn’t old. The name is.

Recently we heard someone say, “If you dislike the mainframe because it’s old, I suppose you won’t buy a Ford either – aren’t they the Model T people?”

The technology isn’t old; it’s as modern as modern gets, and it’s incredible. Even one of the modernization providers acknowledges, “With IBM’s latest launch of the z16 mainframe, IBM has shown that the mainframe is anything but legacy and has an important role to play in the largest enterprise IT environments.” [ii]

After the Spring 2023 SHARE event, Forrester observed

Sessions…reflected a surge in interest in modernizing mainframe environments through the use of automation, DevOps practices, API enablement, containers, AIOps, Zero Trust, hybrid cloud integration, and a range of other topics that you would expect at KubeCon, DevOps Summit, or SREcon.[iii]

Of course, the same blog notes that the z16 is not the right platform for every workload.

Obviously, workloads are moving off the mainframe.

So there is a right way to do it. (Or, at least, there has been a right way for the workloads that organizations have already moved.)

When something seems too good to be true, it frequently is.

There is no shortage of sellers who will tell you that mainframe replatforming is

  • easily achieved,
  • with a quick ROI, and
  • without significant risk.

Marketing materials rarely qualify these claims by scale or complexity.

For example, one promised “current year payback” from a standing start. We’re not here to say that’s 100% impossible. Here’s what we will say:

  • The workload analysis alone frequently takes months
  • This code has frequently been around for generations, supporting the “crown jewels” of the organization, so testing must be exhaustive
  • There is an underlying presumption of security on the mainframe that must either be equaled – or risked
  • And finally, to get a positive ROI in a year requires one “heck of a stroke of luck” on timing.
    • Mainframe licensing agreements don’t work that way. Almost all are multi-year.
    • Almost none scale down linearly with a reduction in workload.

It would be the rare contract indeed that was structured to enable sufficient financial relief to fund analysis, development, testing, and re-platforming efforts in the first twelve months.

There is no substitute for details…

There’s no easy answer. Significant cost and risk accompany any move away from the mainframe.

The questions are legitimate.

During any period of macroeconomic pressure, the mainframe becomes a target. Over the past several years, prominent ISVs have exacerbated the cost issues by bringing high double-digit percentage increases (sometimes worse!) at renewal time. In some cases, they do this late enough in the renewal cycle to make switching impractical before a contract needs to be signed.

Skills sustainability raises legitimate questions about how to support the platform in 2030. There are answers (including us), but they get insufficient attention.

It isn’t easy to evaluate the claims.

There are plenty of skilled sellers out there, some from influential organizations. Perhaps they’ve had success; maybe they are selling. Making the deal can provide short-term satisfaction; finding out what it takes can be a labor of years (and possibly tears).

A few years ago, we worked with a large shop looking to migrate a significant workload to distributed platforms. Our analysis showed that their plan would leave them with – very best case – the same MIPS in 2025 as they had in 2015 – because of organic growth on the portions of the workload that weren’t good candidates for replatforming.

Projects that move MIPS off the mainframe are fine if they bring business benefits and the cost makes sense. But we must evaluate these projects in the context of the workloads that cannot be moved to be sure that the economics work.

Among the stories of multi-year million-dollar projects which disappoint, these two are particularly compelling:

…a Belgian insurer that completely decommissioned its 6,000 MIPS mainframe following a divestiture from a parent company that had a variety of systems running on a mainframe…the CIO repeatedly used the word “nightmare” during the discussion.

The…migration project involved 85 subcontractors and took almost three years to come to the fateful day when the banks switched over. Despite multiple board-level reviews and third-party project audits, the project failed spectacularly and led to chaos for customers, a huge business impact, and a CEO having to step down as a direct result of the failed project and the fallout.[iv]

Gartner wrote that mainframe migrations

typically cost tens of millions of dollars, and once completed, can actually laden the company “with more complexity, more technical debt, and more support issues than when they started…A great deal of analysis must be performed and each cost accounted for”[v] (emphasis ours).

If it were easy, the only mainframe would be in the Smithsonian.

We readily acknowledge that plenty of organizations have moved. If “the low-hanging fruit is gone,” logically, by and large, that leaves the hard work yet to do.

Getting it wrong can cost a CIO their job. Even worse, it can damage a brand: in 2023, IT issues can become customer satisfaction issues which can become corporate reputation issues in the time it takes to type 280 characters.

… and there is no substitute for experience

Unfortunately, this is not the place in this entry where you find “the easy button.” But while it’s never easy, it’s “less hard” with a knowledgeable partner sitting next to you, helping with the analysis, or even asking informed questions based on a professional lifetime of experience with the platform’s architecture.

We can help you vet your proposals: provide feedback and analysis, and subject the workload under study to a fast-paced review of the reality of it working as proposed. Scale is a good example: some selling material we’ve seen recently strongly suggests that any workload north of 10,000 MIPS is a significant challenge for some of these solutions.

We’re not here to tell you what you can’t do. We’re here to make sure you see the inhibitors and build a plan to remediate them

We’ll do it quickly and for a reasonable expenditure. We would never claim to be unique – but we will say that we’re unaware of another consultancy

  • Recognized by industry experts for our mainframe experience
  • Recognized by industry experts for workload placement strategy

So before you go down an expensive, time-consuming and potentially credibility-draining path, talk with a firm that truly understands the mainframe. Specifically

  • We bring true consultative objectivity. We have no financial stake in the outcome- we resell nothing. Full stop.
  • GTSG has been around for over 35 years and on the mainframe every day of it. We understand the environment: a ball of yarn, tightly wound, with each thread color melding into the other in a touchpoint that is hard to see from a distance. We can help you see how to separate the important knots from those you must “cut and reweave” to circumvent.

Thanks for reading this far. If you’d like to discuss any aspect of your mainframe journey, please write Mainframe@GTSG.com and one of us will reach out quickly. Thanks!

[i] https://www.wired.com/insights/2014/04/back-future-cloud-wont-replace-mainframe/, retrieved 04.16.23

[ii] https://www.astadia.com/blog/on-mainframe-modernization-and-hybrid-cloud, retrieved 04.08.23

[iii] https://www.forrester.com/blogs/share-atlanta-2023-preparing-the-mainframe-for-a-hybrid-cloud-world/, retrieved 04.10.23

[iv] https://futurecio.tech/lessons-from-failed-mainframe-migration/, retrieved 04.08.23

[v] https://www.itjungle.com/2019/12/09/moving-off-big-iron-be-very-careful-gartner-says/, retrieved 04.16.23