Oracle Is Nuts To Stick With Sun Hardware

I have to admit, I’m surprised to hear that Oracle really, really does want to keep Sun’s hardware going.

At the time the merger was announced, Oracle made vague noises about supplying integrated hardware and software, but we didn’t believe it. We were sad to see Sun go, but were sure the hardware would be sensibly sold off.

But now, we have Oracle CEO Larry Ellison’s own words that he will be increasing investment in Sun’s SPARC processor line.

“We are definitely not going to exit the hardware business,” Ellison said in a much-quoted interview filed with the US Securities and Exchange Commission. “If a company designs both hardware and software, it can build much better systems than if they only design the software. That’s why Apple’s iPhone is so much better than Microsoft phones.”

He cited Apple again, to back the idea of Oracle making its own processors, and promised to put more money into Sun’s SPARC. “Right now, SPARC chips do some things better than Intel chips and vice-versa. For example, Sparc is much more energy efficient than Intel while delivering the same performance on a per socket basis.”

In our minds, the first question is – why? The answer to that is of course vaulting ambition. Dominating business software isn’t enough, so Oracle wants – and at some level, needs – to do hardware too.

But is this actually feasible? There’s not much long-term precedent for companies doing really well in both software and hardware – especially starting from the level of applications and middleware.

IBM is the most obvious counter-example. It’s still launching new mainframe hardware even now, and doing interesting things with software, including open source. But the overall direction of its hardware is towards standardisation, and greater use of standard chips from Intel or AMD. HP, likewise, does increasingly standard blades, and moves towards software and services. And Dell’s approach has been based on standard processors from the start.

And these are the companies whose strengths in hardware were already beating Sun. What Ellison is proposing, in other words, seems to be to push a lot of money into Sun, give it some protected markets in appliances for its software, and hope that’s enough.

I think he’s nuts. He may be using an up-to-date metaphor with the iPhone, but he’s looking into the past. Servers aren’t iPhones.

IT managers today appreciate the value that commoditisation gives them. Fifteen years ago, it was still possible for a server maker to lock users into its own hardware, but the writing was on the wall, even then.

Users now are chafing under the yoke of Oracle’s licensing, and looking longingly at open source alternatives. They want to minimise their commitment to any vendor, and they want to increase their choices, because they can remember how things used to be.