Worker Shortage May Stymie Strong Mainframe Growth
What Big Iron shops of all sizes have to say about the future of the mainframe is encouraging—to say the least
At last week’s SOA product blitz, IBM Corp. made its case that pervasive service-enablement and the mainframe renaissance are two great phenomena that work great together. Among other encouraging news, Big Blue officials talked up an expected doubling of mainframe-bound transactions over the next four years, growth that they attributed in part to surging interest in SOAs.
Big Iron ISV William Data Systems (WDS) doesn’t have the name recognition of Forrester Research—IBM officials last week cited Forrester’s bullish SOA outlook as one sign among many—but the company does know mainframe customers. What Big Iron shops of all sizes have to say about the future of the mainframe is very encouraging.
Last week, WDS published the results of a survey of 135 mainframe customers across a range of different verticals, including government, financial services, and manufacturing. Nearly one-third (29 percent) of those polled were large organizations with $1 billion or more in annual revenues, while a plurality (46 percent) were companies with between $100 million and $1 billion in revenue.
The big takeaway, says Mary Hegyi, managing director of U.S. operations for WDS, is that the mainframe shops of today are in it for the long haul. In fact, she says nearly half—49 percent—of Big Iron customers expect to grow their mainframe processes over the next six to 12 months.
If anything, Hegyi maintains, mainframe shops aren’t really concerned about the future of the platform; nor are they alarmed by its putatively high cost. Instead, she argues, they’re worried about something else—namely, the comparative paucity of available mainframe talent. In spite of ambitious Big Iron recruitment initiatives, such as IBM’s zNextGen program, many mainframe shops can’t find experienced zSeries or S/390 technologists. In fact, Hegyi indicates, 44 percent of survey respondents said that existing or anticipated worker shortages would delay the deployment of new mainframe-related projects or services.
“The good news is [the survey] basically backs up IBM’s prediction that mainframe processing is going to double over the next few years,” she comments. “About half [of respondents] expect growth in their mainframe processes over the next 12 months. But in order to meet this growth, they’re turning to less-skilled [mainframe] workers.”
That means glad tidings of a sort for the current crop of zNextGen novitiates. After all, she says, by “less-skilled” mainframe technologists, organizations typically have in mind workers with one to six years of Big Iron experience. “This is actually an opportunity for these [zNextGen pros and other not-so-hoary mainframers], because nearly half [of Big Iron shops], 49 percent, are recruiting networking staff with six years or less of experience.”
WDS has a dog in this race, of course. It prescribes its mainframe network monitoring and management products—which include Apias, Exigence, FTPAlert, Implex, and Routeview—as a good Rx (taken collectively or individually) for savvy and not-so-savvy mainframe users alike. WDS’s tools help mainframe technologists quickly pinpoint network-related problems, such as application response issues, transport issues, and the like, says Hegyi.
“[Respondents] consistently said [that] our products allowed less experienced technicians to very quickly identify and diagnose problems, and, secondly, [that] the products themselves were easily implemented,” she claims. “With our FTPAlert product, [for example,] they can see all of the FTP activity that’s taking place in their environment. If there’s a problem, they can see exactly what’s happening so they’ll know how to fix it. They won’t have to hunt through log files or [if they’re mainframe neophytes] hunting for a needle in a haystack.”
Hegyi says the WDS survey helps validate the importance of easy-to-use mainframe network management tooling. For example, she argues, more than one-third (36 percent) of respondents said they expect to experience mainframe network staffing shortages in the next five years. “Programs like [zNextGen] help, but don’t really address the problems in the near-term,” she argues.
For example, Hegyi says, 42 percent of respondents cited application failure as the most common source of mainframe problems, with network failure coming in a close second at 36 percent. With so many organizations undertaking network-dependent, service-oriented application development, this should be a real area of concern for mainframe shops. “There’s a growing shortage of skilled technicians to address these issues. It’s a problem now and it’s going to be an even bigger problem down the road, even with all that IBM is doing. Many organizations are going to be put in an uncomfortable spot when they’re deferring deployment of new technologies for want of knowledgeable network staff.”
Elsewhere, the WDS survey revealed that 71 percent of respondents currently run their mainframe applications on a mix of network protocols (SNA and TCP/IP), and that 34 percent expect to continue doing so. The reason, according to most respondents, is the prohibitive cost of rewriting existing SNA-based applications. Approximately one-third of respondents are currently running an all-TCP/IP environment, Hegyi says.
Stephen Swoyer is a Nashville, TN-based freelance journalist who writes about technology.