In-Depth

Q&A -- Looking Back, Looking Ahead at the Mainframe

IBM's zSeries marketing manager reviews the mainframe's accomplishments, and explains how mainframes are being deployed today.

• By Stephen Swoyer
• 03/25/2004

Existing customers are adding capacity at an encouraging rate, Mastrobattista confirms, and most of the capacity that Big Blue is shipping (on its new T-Rex mainframes and elsewhere) is earmarked for new workloads—an encouraging reversal of a historical trend. Plus, with Big Iron Linux providing a natural entrée for many of today’s Linux-proficient college graduates to the mainframe world, Mastrobattista says that the mainframe’s future has never looked better.

[Last year] was a big year for Big Iron, with the announcement of T-Rex [the z990] back in May, along with several other prominent events. Could you give us a recap of some of the other things you you announced last year?

Don’t forget that we also had the Mainframe Charter, which we announced in August. That was an attempt to show the marketplace IBM’s commitment to the platform, and that the zSeries platform was going to remain a strategic platform in the IBM eServer line.

We followed that with the October GA2 announcement for z990, with some additional features, we went from 15 LPARs to 30 LPARs, we announced the availability of the larger C and D models, with 24 and 32 processors respectively. Then we also announced our Infrastructure Simplification projects, which is really this idea of the mainframe being a key part of projects where our customers are trying to simplify what has become a complex infrastructure. So we married this idea of mainframe technology with blade technology, which was not the most intuitive idea out there in the market place, but which some customers are already doing.

When you look at that whole drum beat of announcements, the fact that we delivered on something very innovative—which was the z990 platform—along with a resounding commitment from IBM to tell customers that zSeries is a platform they can depend on for the future.

Is that what you’re hearing from customers? That Big Iron is a platform they feel they can depend on for the future?

I think what people are doing today is they’re really looking at some of their key pain points, and the key business pain points today tend to center around financial pressures, simplifying the infrastructure complexity, deploying new capabilities for both business as well as an IT perspective. I think people are realizing that the mainframe has really been around now for 40 years now, that the mainframe has been out there, and the mainframe has always been designed for mission-critical workloads. What that means is that more and more customers are realizing that the mainframe is the platform that they need to deploy their mission-critical workloads. All of the things that the mainframe was good at doing—systems management, performance management, security—these are all core values that the mainframe has been very, very good at. So [customers are] saying, “Gee, this is a platform that has been really good at doing these things, so why should we use anything else?”

What kind of feedback are you getting from customers with respect to their plans to cut, maintain, or increase their mainframe capacities? And do you have a breakdown of sales in terms of mainframes slotted as replacements for existing systems and mainframes brought in to support new workloads?

The feedback that we’ve been getting, especially after everything we announced last year, has been very encouraging. Many of our traditional customers are buying more capacity, so we’re still selling a tremendous number of MIPs to these traditional customers. That’s kind of a core business element to today’s enterprise. As for new workloads, the best way to answer that—when we look at 1998, which was now about six years ago, at that point, only 15 percent of the MIPS that we shipped would have been in support of new workloads, so the remaining 85 percent would have been dealing with the more traditional online transaction processing, like CICS. That percentage is now well in excess of 70 percent, so in six years time, people are really looking now to the mainframe to be a more strategic part of what they’re now doing.

While companies might be maintaining or augmenting their mainframe capacity, are they making any effort to train additional IT staff on the mainframe? Is there anything that IBM can do to help them with this?

Well, the fact that Linux is on the mainframe, that means that it now taps a much different skill set in the marketplace. More and more young professionals graduating from college have more of a Linux background than they would a System 390 background. But the fact that the mainframe has really evolved to encompass more strategic workloads helps to address some of the skill set issues. If you look back to the Mainframe Charter that we announced last year, there were really three elements to that charter: one was innovation, one was value and one was community. We’re trying to enrich the zSeries community, and that involves university programs. That involves helping customers grow their own mainframe skills sets, so we’re evolving more of our application architects into a zSeries environment, helping to ensure, again, that zSeries is a platform they can depend on in the future.

You mentioned Linux and the Mainframe Charter. Another aspect of that was standardized [Integrated Facility for Linux] pricing to make it more affordable for more customers. What are some of the changes that you made in this respect?

What we did there is, we wanted to make some pricing of the mainframe world more visible to the marketplace. One of the areas that we did that was to standardize IFL pricing across the board, so whether a customer has a G5 server, a G6 server, a z900, or a z990 server, the price of the IFL engine was announced as $125,000. What that does is inherently provide the newest generation hardware the best price/performance, because as the engine capacity gets larger, we’re still providing that increased engine capacity at the same $125,000. We made the price tag visible, but we also made it in a way that going forward, the newest generation tech always has the best price/performance.

I wanted to switch gears for a moment and talk about one aspect of mainframe cost of ownership, which is part of the price/performance you were talking about: capacity pricing. One of the concerns I’ve heard from customers has to do with sub-capacity pricing, which IBM meant to tackle with a License Manager tool that you had on the drawing board some time back. Can you say anything about whether this will make an appearance at some point?

We do have a sub-capacity reporting tool that’s out there right now, and that tool is designed to really help customers better gauge and also predict what the right capacity should be for some of their software products. The IBM License Manager concept was a concept that was originally going to bridge or span multiple eServer platforms. What the company had decided to do was to focus initially on delivering sub-capacity and the tool for delivering it on the zSeries. What we have instead is a sub-capacity reporting tool that runs exclusively on zSeries, and it’s designed to help the customer. It tells the customer what percentage of processing capacity is used, and that’s strictly a zSeries tool.

Finally, you’re gearing up to celebrate the mainframe’s 40th birthday next month. Any thoughts on the platform as it notches this important milestone?

In April of 2004, we’ll be celebrating 40 years of the System 360, and in those 40 years, the mainframe has really had a series of core values that it has been very well noted for. These are all core values that have helped differentiate the mainframe over time, and they have given the mainframe the ability to call itself an industry leader in these core values.

What’s happening today is that the amount of complexity that has evolved into the everyday IT infrastructure has allowed people to say, “Hey, I need to simplify this complexity,” and they’re going back to those core values and using the mainframe as an infrastructure simplification platform. I’m not saying that the mainframe is the one-size-fits-all platform, because we at IBM don’t have a one-size-fits-all platform. But people are finding that the mainframe is very critical for helping simplify people’s IT infrastructures going forward.