Oracle To Sell Sun's Hardware Business To HP?

Underholdning writes "With the DOJ approving Oracle's Sun buyout, the question arises what Oracle might want to do with Sun's hardware business. It's no secret that what Oracle wanted was the software part. Now The Inquirer is running a story claiming that Oracle will sell the hardware business of Sun to HP. This will give Oracle a juicy check while HP can increase its services. Larry Ellison denies that it will take place, but a source for CNN claims otherwise."

Comments

Sparc and Solaris

[Powys]

Since Sun made Sparc cpus, and used Solaris for their OS, will they sell Solaris to HP to match their hardware?

Not happening

[the linux geek]

HP had a hard enough time last time they tried to support multiple processor architectures simultaneously (for a while, they were selling x86, PA-RISC, Alpha, and Itanium.) I don't think they're that interested in adding Yet Another OS and Processor Combination into the mix, or they wouldn't have axed PA-RISC and Alpha, both of which had real futures. They've implied strongly in recent times that they're committed to Itanium, and I think that's where it will stay.

What does Oracle want from Sun?

[Futurepower(R)]

If Oracle does not want Sun hardware, what Sun software does Oracle want?

My theory about why has Sun Microsystems not done particularly well in the last few years is that the highly reliable hardware Sun Microsystems sells is no longer popular because it is far cheaper to use consumer-grade hardware with software that is fault-tolerant. The excellent 2008 book Planet Google describes Google's experiences on page 54: "For about $278,000 in 2003, [Google] could assemble a rack with 176 microprocessors, 176 gigabytes of memory, and 7 terabytes of disk space. This compared favorably to a $758,000 server sold by the manufacturer of a well-known brand, which had only eight multiprocessors, one-third the memory, and about the same amount of disk space."

It's true that Sun hardware is more reliable than consumer-grade hardware. However, neither are completely reliable. Both require fault-tolerant software. Also, consumer-grade hardware has become very reliable.

Re:What does Oracle want from Sun?

[blind biker]

I am sure Google made the right decision. I am also sure, however, that an 8-way MP computer with loads of contiguous RAM will excel at some tasks where a cluster will not. A cluster is not always a direct replacement for a Sun M8000. Each has its uses and its rightful place in IT.

This is about buffing up HP's

[idontgno]

IT services arm.

The Inquirer (the IT news website, not the tabloid) has some words about this:

HP, on the other hand, wants Sun's hardware to boost its services business. HP bought outsourcing player EDS. EDS was Sun's best customer. By owning Sun technology, HP will improve its profit margins on many EDS deals.

HP told CNN that the EDS integration process has gone well and the subtext is that the maker of expensive printer ink is fine with writing a big cheque to Ellison.

So, HP bought EDS, and EDS has a historical habit of recommending or BOM'ing Sun hardware. Solution? HP buys and manufactures Sun hardware. That way, EDS is eating HP's own dog food. That's the "x) PROFIT!" stage.

Re:Talked to Sun guys yestorday

[glwtta]

sun would as a hole be run as sun a Oracle company idea as a hole

Could you do me a favor? Just take a minute to read that sentence to yourself three or four times. Maybe even read it out loud - people say that helps.

HP buying the SPARC hardware biz -- seems unlikely

[twasserman]

While I think that it makes good sense for Oracle to sell off the hardware business that they acquired from Sun, I find it hard to imagine H-P as a potential buyer. I'd be much more likely to believe the story if the buyer were Fujitsu, who has been involved with Sun and SPARC since the 1980's.

Here's why I think that H-P is unlikely to do this:

• They are trying to focus more on their services business, having recently spent $14 Billion acquiring EDS

• The SPARC line and the Solaris operating system would go head-to-head with HP's high-end servers (Itanium-based) and HP-UX operating system. While it would take a strong competitor off the table, it would also create uncertainty with the large SPARC installed base, especially in the financial community.

• H-P's acquisition of COMPAQ wasn't exactly a roaring success. The only good thing that came of it was that it led to the downfall of Her Worship, La Fiorina, who viewed the acquistion as her crowning achievement, even though it led to firing 30000 people. Today the COMPAQ brand is fading away and mostly is used for inexpensive PCs.

• Mark Hurd is extremely cost-conscious and very focused on quarter-to-quarter results for Wall Street. A big acquisition like this would be very disruptive and require some significant writeoffs and future earnings impairments.

I would not be at all surprised to learn of talks between Oracle and HP, but I would really be shocked if this deal happened.

itanium

[turgid]

What would HP want with that old SPARC junk when itanium is quite clearly the future?

Re:Hey, why not?

[ThePhilips]

No no. You got it wrong. HP's model is "buy it, burry it, write off losses."

Frankly, having seen what HP did to DEC (acquired as part of Compaq; the ill-fated Alphas and Tru64), acquisition of SPARC and Solaris would spell rather quick death to both. In its current shape, HP unlikely to be allowed to do the trick again. (Nor Solaris customers would want to migrate to HP-UX, which is probably most POSIX-incompatible POSIX-certified OS I have seen to date.)

As a UNIX seller, HP is probably most backward company you can find out there. And their upper management who are forgetting at times that they still have UNIX business doesn't help to improve the image.

Oracle is advertising Sun hardware

[azrael29a]

Why in hell would they want to sell the Sun hardware business to HP when they have just started to advertise it? See http://www.oracle.com/features/sunoraclefaster.html

HP, where RISC goes to die?

[damn_registrars]

As someone who has used both Alpha and SPARC chips in high-performance computing environments, I was a bit saddened when the Alpha went away for good. Seeing the SPARC also go the way of the Dodo would be a shame as well.