The Mainframe's Youthful Injection
Last spring, one of my co-workers went to college campuses to recruit prospective "young mainframers."The death of the mainframe is a perennial conversation topic for IT pundits. Not only is it still alive and kicking, but young developers are being sought to extend its lifetime.
Young mainframers? Isn't that an oxymoron?
My co-worker, Tim, explained that our company, a major software vendor, is seeing its mainframe workforce rapidly approaching the age of retirement. Tim said IBM and most other firms whose businesses depend on mainframes are also dealing with this industry-wide problem.
Since the 1980s, PC's and UNIX machines were supposed to have taken over the computing world, relegating mainframes to the scrap heap alongside rotary-dial telephones, suitcase-size boom boxes, and Plymouth Reliants. Indeed, most mainframes from that era have been consigned to the scrap heap - only to be replaced by bigger and faster mainframes.
Today, the number of mainframes is estimated to be 10,000. Since 2000, the processing power of mainframes has quadrupled in terms of MIPS. According to IBM, the top 25 world banks run mainframes, 80% of the world's corporate data resides or originates on mainframes, and 71% of global Fortune 500 companies are mainframe clients.
And what about those applications that were supposed to have been migrated off mainframes?
As one mainframe veteran put it, "We started sun-setting some of our mainframe systems so long ago, the sun's rising again!"
While reports of the death of the mainframe "are greatly exaggerated," to quote Mark Twain, the same cannot be said about the mainframe workforce, whose average age is measured in minutes from retirement.
To replenish the thinning ranks of the 3270-keyboard jockeys, companies have started recruiting young talent to be trained in the technology of the mainframe. They're counting on an 'Inter-Generational Transfer of Expertise' from old mainframers to young mainframers before the sun sets on those old mainframers.
I caught up with one of these young developers to get a sense of what enticed a new college graduate to cast his lot with the venerable mainframe.
Francisco Esqueda graduated from The Pennsylvania State University with a computer engineering degree in May, 2009. His education included development languages like Ruby, Java, and HTML, among other technologies, and he honed his skills with summer internships working with web applications and PCs.
Although he "had no real concept of what developing for the mainframe was like," he accepted a position as a mainframe developer.
"My mother had reservations at first," he admitted. "She was shocked to see that I would be working with the same languages and tools that she had started with in her career: Assembler language, REXX, JCL, ISPF. She was worried that working with 'that stuff' would be a career dead end."
Francisco comes from a family where conversations about bits and bytes were as much a part of dinner as bread and butter-or as much a part of dinner as Apple and spam (to keep the prose computer-related).
His mother retired in 2000 after a 25-year career in IT, during which she became CIO of a large oil company. His father, a Ph.D. in electrical engineering, is an associate dean at the The Pennsylvania State University, Berks campus. His older sister and brother are also degreed electrical engineers.
According to Francisco, his mother would have never predicted that the mainframe 'stuff' that she left behind ten years ago would be the foundation of his budding career.
"She changed her mind after she researched it," he continued. "She learned that companies are still committed to the mainframe, and there were new initiatives. And she liked the training program I was offered."
Francisco recently completed a two month training program that covered the z/OS operating system, Assembler language, DB2 database, VSAM file structure, and CICS ("kicks" to those in the know), among other topics.
How is the 22-year-old Francisco, a child of the cell phone-toting, Facebook-connected, point-and-click generation - with no exposure to mainframes prior to June of this year - handling mouse-less keyboards with 24 PF keys, dataset allocation, and core dumps?
"The interface is not good, it's cryptic," he said. "It takes a while to get the hang of it."
On the plus side, he sees the low overhead of the interface as an advantage. "You don't have to wait for some IDE to load. The mainframe is really fast. And with Assembler language, you have complete control over the efficiency of your programs."
As far as he knows, Francisco was the only Penn State computer engineering graduate of the 2009 class to take a job working with mainframes. Most of his fellow graduates ended up working in Web-centric systems. After graduation he joined 20 other recruits, most of them recent graduates, in the mainframe software engineering training program of a large software vendor.
Did Francisco feel compelled to take the mainframe job because of the state of the economy? In other words, was the mainframe job the only show in town?
"I had two other offers," he explained. "One was working on a UNIX Web server but I liked the training program and the idea of working for a software company."
Several companies that depend on mainframes obviously think training young mainframers can help resolve the mainframe skills shortage. Francisco seems to agree with them, and is apparently convinced that there's a future in working with the platform of big iron. In September he returned to his alma mater, Penn State. He went there to recruit "young mainframers."
Although Edward J. Joyce no longer logs onto the mainframes at CA,, where he is a principal software engineer, he is still a young mainframer at heart.
Article courtesy of Datamation