In-Depth
Mainframe Pros Optimistic about the Future
Mainframe pros sound off about the year ahead
In retrospect, 2006 was a pretty good year for mainframe boosters. Mainframe revenues surged and IBM Corp. unveiled a raft of Big Iron-centric initiatives, including a new specialty engine (the zSeries Integrated Information Processor (zIIP)), new mainframe hardware (its z9 Business Class and z9 Enterprise Class systems), a new version of z/OS, and a $100 million, five-year investment in mainframe simplification.
For these and other reasons, many Big Iron professionals say they’re optimistic about the year ahead. That doesn’t mean, however, that they anticipate completely smooth sailing.
“I believe introduction of the z9 machines and investments in related new technology were essential to address the roadblocks that customersfound when trying to run new workloads on their mainframes,” says Rob van der Heij, a mainframe technologist with VM performance management specialist Velocity Software Inc. “Combined with the strong systems management in the mainframe area, we have a very competitive platform.”
That’s the good. The not-so-good—but not-altogether-bad, of course—is that cost-friendly options such as zIIP (or any of several other software specialty engines IBM offers) complicate capacity planning and management on mainframe systems, Van der Heij argues.
“These specialty engines so far are primarily a software license tool for z/OS installations and may be an attractive solution for those problems,” he says. “In general splitting up your processing capacity into units with specific affinity makes it harder to do capacity planning and tuning. There's a good and a bad side of that, but I don't expect IBM to do these things for z/VM as well, so it does not bother me yet.”
This concern notwithstanding, van der Heij—like other Big Iron boosters—says he’s heartened by the mainframe’s recent winning streak. However, many mainframe technologists still feel it’s an uphill battle for Big Iron recognition and mindshare.
Consider the case of Stephen Frazier, a mainframe pro with a state government agency based in the Southwest. Frazier’s organization once planned to quit the mainframe entirely. That was one executive change ago, however, Frazier says. His new boss seems more open to the mainframe, at least insofar as it can be a robust platform for Linux. As a result, he says, his organization plans to cautiously expand its mainframe efforts in the coming year.
“We are adding new functions to existing mainframe applications this year. We moved one application off the mainframe in 2006. We are not planning to move any off in 2007. The previous management—now retired—wanted to move everything off the mainframe but kept having problems getting the replacement applications to work,” he indicates. “The new manager allowed the one application move that was in progress to be finished but has not started any more.”
Frazier isn’t sure what this means for the mainframe’s future in his organization: “No direction has been set, but he is from a Unix background and has hinted that he thinks Linux should replace Windows wherever possible. I don't know where he thinks the mainframe is going to fit in.”
The Price Still Isn’t Right
Frazier, too, seems heartened by some aspects of IBM’s recent mainframe push. At the same time, he notes, Big Blue is still missing the Big Picture. Mainframe hardware—even a $100,000 mainframe system, such as the z9 BC—is still too expensive for most SMBs. Only commodity-based mainframe solutions—e.g., emulation systems such as FLEX-ES from Fundamental Software Inc.—can really compete in this market. Yet IBM refuses to make an important FLEX-ES-related concession, Frazier argues.
“IBM's refusal to allow 64-bit systems to run on commercial FLEX-ES machines is one of the worst and most short-sighted decisions in IBM's history,” he contends. “This means that small businesses will never be able to use mainframes. As small businesses grow into large businesses, they will not be IBM mainframe customers. IBM's moves in 2006 (if not reversed) will lead to the destruction of the company.
Elsewhere, cautious optimism abounds. David Wakser, a mainframe pro with a prominent outsourcer based in the U.S. Northeast, says that while his company does offer mainframe outsourcing services, it’s on a demand-only basis. To Wakser’s knowledge, his employer isn’t actively increasing its investments in mainframe hardware or software—nor is it likely to do so.
“The company I work for is a major outsourcer, which means its mainframe needs—or the events that drive its mainframe needs—[are] entirely client-based and in no way dependent on us,” he indicates. “As long as clients want to outsource their needs for a mainframe and our company can make money from that need, our company will continue to retain [its mainframe assets].”
Why isn’t Wakser’s employer expanding its mainframe investments, at least in view of demonstrable demand among clients? For the simple reason, he says, that few clients are outsourcing next-generation workloads. “[My company] would only look towards newer mainframes if the clients required features there that were unavailable on older models. Otherwise, they can provide the same services at a cheaper price by staying with slightly older technology.”
The flipside to Wakser’s case is that companies tend to outsource applications or services that they aren’t actively growing in-house. As a result, application maintenance—particularly of legacy applications—is one popular outsourcing workload, at least in the mainframe space. IBM officials, for their part, cite surging growth in zLinux, WebSphere-on-z/OS, and other next-generation workloads, and most of these workloads are not only staying in-house, officials claim, but getting deployed on new System z hardware.
“There was a 25 percent year-to-year revenue growth in the mainframe, and we had 18 percent year-to-year MIPS growth” in the third quarter of 2006, said Bob Hoey, worldwide vice-president of System z sales for IBM, in an interview late last year. The big driver, Hoey claimed, was a mainframe value proposition that simultaneously (and inexpensively) encompasses cutting-edge workloads—such as zLinux, zIIP, and WebSphere (via Big Blue’s zSeries Application Assist Processor, or zAAP, engine)—and traditional VM, VSE, and z/OS applications, too.
“We are seeing just enormous interest in [mainframe] Linux, in addition to WebSphere and the new zIIP processor, which as you know just became available a few months ago [in June of 2006],” Hoey concluded.