Mainframes & COBOL: Setting the Record Straight

Overview: As some states struggle with a surge in unemployment claims, the mainframe is being unfairly criticized as an outdated platform. The fact is that the world’s economy relies on the strength and reliability of the mainframe—and its relevance is growing.

These last few months have tested our mettle in almost every aspect of our lives. I have the good fortune to be able to work from home during this time, but many others—fellow Floridians—are not so fortunate. I deeply sympathize with those whose unemployment checks have been delayed, but as a technologist, I feel like I need to set the record straight on COBOL and the mainframe, as they are being wrongly blamed for the struggles some U.S. states are having fulfilling unemployment applications.

I want to first start with highlighting an article from the Orlando Sentinel dated May 21, 2013. It’s important to try and understand how we got here and why I feel compelled to write this blog. The headline reads: “State to finally replace ancient jobless-claims computer.” Here are a few excerpts:

In good news for the jobless and employers alike, the state’s 1970s-era computer that processes unemployment claims is finally getting replaced.

The overhaul will replace an aging—and now infamous—mainframe computer system left over from the Nixon era. Installed in 1972, the mainframe and network it supports have been modified repeatedly through the years, and the mainframe has become increasingly unreliable and difficult to program.

A report warned that the mainframe was “well beyond its useful life” and at an “ever increasing risk of … failure.” A separate report done a year earlier had identified a dozen “major inefficiencies,” affecting everything from client intake to claims appeals.

The replacement of the new platform is being paid for with federal tax money.

That’s one reason why many states have operated for decades with out-of-date computer systems. A 2009 Maurice Emsellem, co-director of the National Employment Law Project, said that, traditionally, Congress has put little focus on helping states improve the administration of their unemployment programs.

This would prove a fateful error.

Fast Forward to April 2020

Today, Florida’s unemployment system called CONNECT —an open systems platform that replaced the mainframe some eight years ago—is now in the news for failing to support hundreds of thousands of Floridians attempting to file for unemployment benefits. An April 15 Guardian story paints a heart-rending account of people trying desperately to file for benefits only to be kicked out of the system multiple times.

Hindsight is indeed 20/20, but does anyone reading this wonder how the world’s largest banks, insurance companies and airlines have been able to thrive for decades using mainframe technology that was unfairly and incorrectly labeled as “well beyond its useful life”?

For those who aren’t aware:

  • Mainframes run 30 billion transactions per day, hold 80 percent of the world’s business data and handle 90 percent of all credit card transactions.
  • Mainframes host more transactions daily than Google.
  • The new IBM z15, is configurable as a one to four 19-inch frame system, easily aligns with the modern cloud data center.

Does this sound like a Nixon-era computer?

As an ITProPortal article aptly points out, “New technologies are constantly being introduced into our lives only to be usurped a few years later, but amidst all the toing-and-froing of the past few decades in large enterprise IT, there has been one constant: the mainframe. Despite false claims of the death of the mainframe, it remains the bedrock of many organizations, powering high-volume transactional workloads and mission critical business functions. The mainframe’s strategic relevance is only growing in the digital age; in fact, 64 per cent of mainframe-powered organizations are now planning to run more than half their mission-critical workloads on the platform, an increase from 57 per cent in 2018.”

Taxpayer-Funded Band-Aids

Last week, the state of Florida rolled out measures to control the flood of applications. The state added 72 servers to allow more people to be on the website at once and introduced a second, mobile-friendly site. Governor Ron DeSantis said that 750 new call operators were deployed to take calls.

How much is this costing taxpayers? Contrast that to a study conducted by Howard Rubin, CEO and Founder of Rubin Worldwide, that found that while computing power has doubled over the last five years, server-heavy organizations’ costs have gone up 63 percent more than mainframe-heavy organizations. Mainframes account for 68 percent of production workloads but only six percent of IT spend, with their average IT costs of goods being 35 percent less than server-heavy organizations.

The world’s economy rightfully runs on more than 220 billion lines of COBOL code, and that number is increasing every day. Just to make everyone aware the current version of COBOL is V6.3. Why is this important? COBOL is continuing to be developed and maintained regardless of what people are saying.”

A Silver Lining

There are no short answers and plenty of blame to go around for past decisions that are now resulting in abysmal consequences.

My hope is that state CTOs and CIOs will realize the incalculable value they have in their mainframe data and applications. Beyond that, it’s incumbent upon them to provide a modernized development environment so developers of all skill levels can update and manage COBOL applications as seamlessly as they do other applications. Essential to this is an eclipse-based IDE for mainframe development like Topaz to help them build, test and maintain the applications on the mainframe.

I want to end this blog on a positive note. The Governor recently remarked, “The [unemployment] system just totally broke. It’s not a good system. We’re going to deal with that.”

Progress is being made already.

This post originally appeared on LinkedIn.

The post Mainframes & COBOL: Setting the Record Straight appeared first on Compuware.

Read the original article

Related posts

Leave a Comment