This can't be an easy time for Ann Livermore. When Hewlett-Packard Co. missed Wall Street's earnings expectations late last week, the blame was placed squarely on the shoulders of the Enterprise Servers and Storage Group, one of the divisions she manages. "Unacceptable" problems within the group cost HP $400 million in revenue and $275 million in operating profit, said HP Chairman and CEO Carly Fiorina just hours before HP announced the sacking of three senior executives within the division (see story).
Though HP says the operational disruptions that caused last quarter's shortfall are now behind it, daunting challenges remain for Livermore, who, as executive vice president of HP's Technology Solutions Group, manages the company's enterprise storage, systems, services and software products. Livermore must pull off the trick of moving a diverse set of PA-RISC, AlphaServer and HP 3000 customers to HP's Integrity line of Itanium servers, while simultaneously shoring up HP's once-proud storage business, which has been struggling of late.
Livermore, a 22-year HP employee who was considered a front-running candidate for the job that went to Fiorina, met with Computerworld's Patrick Thibodeau and Robert McMillan of the IDG News Service at HP's annual HP World user conference in Chicago this week to discuss the state of her company.
Why did the enterprise computing group lose money this quarter? There were really a couple of main areas where we didn't execute as well as we needed to this quarter. In the United States, we were transitioning to a new system and set of business processes for our order management and manufacturing operations, and that transition caused more disruption than we had planned. As a result of that, we didn't ship as many of our servers, particularly our ProLiant servers, as we had planned during the quarter.
In Europe, we had some issues around our channel management processes, as the European team centralized some of the claims processes and channel compensation processes. So that was another area where we just didn't execute well. We also had, in a few instances, in EMEA [Europe, the Middle East and Africa] more aggressive discounting than what we had originally planned for.
So the good news -- or the bad news -- is that they are all internal execution issues that we can address ourselves. They're not things that are fundamental weaknesses in our business offerings.
What about the storage group? With the storage business, we introduced a set of new product offerings back in April, which was a strong refresh of the product line, and we have another major set of announcements in September, so we feel good about our product offerings. We are at the point where we want to put in place some additional sales specialists to help in our storage sales, and also we have a few of our value-added resellers that we're working with to try to boost ... the selling of our products.
You're getting rid of the Alpha chip, and you're moving from PA-RISC to Itanium. Is there any sense that the decline in sales may be indicative of customers just not wanting to move to Itanium? No. What we're seeing with Itanium is that our quarter-to-quarter ramp is as we've expected, and our Unix growth last quarter was 8%, which is a good year-over-year growth rate.
When you say that Unix growth was 8%, do you mean on Itanium? Unix growth overall.
So, how was it doing on Itanium then? You know, I didn't bring with me the specific breakdown in terms of what the PA-RISC growth was versus the Itanium. But the thing that we're seeing that we like from an Itanium perspective is that the ISV [independent software vendor] applications and ports are going really well. Our ISV partners have all believed that the migration has been easier than what we planned.
Did Itanium sales hit your expectations last quarter? They did. The Itanium sales did, and our overall Unix growth hit our expectations as well, so we performed well in that product line. And that's important for us, because although Unix is a slower overall growth market, it goes in very mission-critical applications that tend to sell a lot of services around our Unix environments, very often storage as well. So that's a very important operating environment for HP.
IBM is sticking with its Power architecture, and Sun Microsystems is sticking with UltraSparc. Is there anything in these results, or in feedback from customers, that is giving you any grounds for doubt about your Itanium-only strategy? No. And we don't have an Itanium-only strategy. From a customer perspective, we have a very simple set of architectures: We have Itanium, and we have x86. A lot of the power of HP from a computing perspective is we do offer both. We can offer our Linux or Windows and our Unix offerings on Itanium, and we can offer Linux and Windows on our x86 boxes.
Has Intel's embrace of the 64-bit extensions to x86 affected your x86 strategy at all? Would you think of porting HP-UX, for example, to the x86 platform, now that there are these 64-bit systems? No, we don't think that we need to do that. For us, it's just another great offering that we can bring to the market at a very low cost because of the presence and familiarity that we have with the x86 architecture.
So, what do you think of Sun's effort to promote Solaris on x86? They're pretty late to the game in terms of the offering. Customers look to what executives say over the long term and don't always take it well when you change your strategy or your tune. They waited very late to make that step.
There's been an ongoing push to get the HP 3000's MPE operating system open-sourced. What's your position on that? We've not taken any kind of position or decision around doing that. We're at the point where we're still working very aggressively with our customers to help them with their migration activity, and they're working against our 2006 timeline. At least to date we feel good that we have got mechanisms, approaches, programs, all those kinds of things to work with our customers.
Would you consider open-sourcing HP-UX in the way that Sun is now promising to open-source Solaris? That's not a choice or priority that we have, and the basic reason why is we have a tremendously high customer satisfaction with HP-UX that's running in very mission-critical areas. We've got great services and support capabilities around it. When we look at the applications and the nature of it, we don't believe that there is a compelling business reason for HP to do that.