I’m starting to think it might be time to ditch the mainframe; the name, that is, not the platform itself. This may sound strange coming from someone who spends so much time extolling the virtues of continuing to use legacy code and mainframe processing as part of modern hybrid systems that meet today's computing needs. I think it’s the term “mainframe” that seems to be holding us back and stopping us from getting away from the ”it's old, let's ditch it” mentality.
This is not an entirely new idea; in fact it has support from several experts, including respected IT industry analysts such as
Ovum and Quocirca. Having done so much to highlight and overcome the stigma attached to the term ”legacy,” as applied to the mainframe- based application code that still supports a vast amount of today's business processes, it seems we now need to turn that same attention to the term “mainframe” itself.
The basic logic is simple: because the term mainframe has been coined for so long, anything that is called a mainframe is perceived by today’s IT generation to be outdated technology–unsuitable for the demands put on today’s systems. Of course that's not true; while the mainframe was introduced 50 years ago, IBM has maintained a relentless pace of innovation to evolve the platform. Think about the zEnterprise EC12 of today; it uses the world’s fastest commercial processor, incorporates new solid state memory technology and supports private cloud environments by consolidating Linux workloads. It is the product of more than $1 billion in R&D spend, but is still commonly referred to as a mainframe!
The truth is the mainframes of today are so far in advance of (and removed from) the mainframes of yesteryear, that it really would not be too far-fetched to call them a totally different animal; but what animal should that be? The term mainframe seems to devalue what all of the new z/OS enterprise computers that IBM keeps introducing are really about, in the same way the term legacy has negative connotations when it's applied to application systems. We helped on the legacy front by talking about an application heritage, but what can we do for mainframe? One commentator has suggested, “Call it anything—call it Jennifer if you want—just get away from mainframe,” and concludes that, “IBM has got to change the way it is messaging these items.”
For more mature IT professionals, the term IBM mainframe conjures up perceptions of high reliability, power and quality, along with positive emotionally charged attachments related to the fact that the mainframe has probably been the primary computing platform ever since they entered the industry. But for newer IT workers, who are growing up with mobile and tablet devices, that is not the case; they don't share that long-held respect for the mainframe’s power and reliability.
So, while IBM might now be building mainframes that are perfect for running Linux as well as z/OS workloads and for hosting complex hybrid systems on a single box, the people who buy Linux servers today are often turned off as soon as they hear the word mainframe. They never get to experience the real value of the platform.
The question is: should IBM’s marketers invest their resources in trying to fight and change these negative perceptions, or would it be more effective—and easier—to start fresh and simply call newer mainframes something else? Is it easier to get people to accept a new idea, than to get them to change their minds about something they already believe?
So what we need now is a new name for a totally different animal; and that's the next problem: what do you call it? Jennifer?
Philip Mann is Principal Consultant at mainframe performance management expert, Macro 4. Philip has been working with IBM mainframes for in excess of 30 years, including over 10 years with Macro 4, where application performance tuning is one of his major interests.