Oracle (Nasdaq: ORCL) has a long track record of making tough software acquisitions work. But in taking on stumbling, money-losing hardware ex-giant Sun Microsystems, it's starting to look as if the company bit off more than it can chew.

Plummeting market share
It might be true that Oracle acquired Sun more for its software than its hardware; the company's Java programming language and MySQL database platform were especially prized assets. But all the same, Oracle needs to make Sun's once-dominant UNIX server business a success to justify the $7.4 billion price tag attached to the acquisition. Critics of the deal noted from the start that this wouldn't be easy, given the steady market-share declines that proprietary UNIX systems had been seeing over the past decade at the hands of cheaper systems running Linux and Microsoft's (Nasdaq: MSFT) Windows, and powered by Intel (Nasdaq: INTC) Xeon and AMD (NYSE: AMD) Opteron processors.

These fears only got stronger in the months immediately following the deal's announcement, as customer uneasiness over the deal's impact on Sun sent its market share into freefall. UNIX archrival IBM (NYSE: IBM) enjoyed particular success in poaching Sun clients.

Meanwhile, Gartner first-quarter server market share data, released earlier this week, threw some cold water on the hope that Oracle's closing of the Sun acquisition in January would help stabilize the situation. Gartner estimated that Sun's Q1 server revenues fell by 38.7% annually, leading its market share to plummet from 9.6% to 5.6%. Even worse, sales of "x86" systems, largely based on Intel and AMD's Xeon and Opteron processors, were estimated to grow by 32.1% over the same time. Strong demand for x86 hardware benefits Hewlett-Packard (NYSE: HPQ) and Dell (Nasdaq: DELL) -- two companies that can't be thrilled that a longtime partner such as Oracle is now a competitor.

Does Oracle have a clue?
Beyond Sun's ongoing market-share collapse, Oracle's recent actions and comments suggest that the software colossus is in over its head trying to turn around a large hardware business. Already, the company has angered Sun customers and resellers by drastically increasing the price of customer-support services for Sun's servers, and altering the general structure of support contracts.

It looks as if Oracle thought it could get away with this maneuver because it had succeeded in doing the same with software acquisitions. But instead, it's learning the hard way that -- surprise, surprise -- the hardware business is a very different animal, and running it like a software business can trigger a strong customer backlash.

Likewise, recent comments from Oracle CEO Larry Ellison in a Reuters interview suggest an obsession with improving Sun's bottom line over the short term, at the risk of further damaging the long-term viability of its server business. Ellison bemoaned how Sun lost money on some of its deals and also took a loss in reselling products from Hitachi and Symantec. Ellison's used to the sky-high gross margins enterprise-software companies see on almost every deal, and his opinion suggests a lack of understanding that hardware companies sometimes need to take a loss on deals to maintain an account that's still quite profitable overall. Insist on never taking a loss on a transaction, and you might have trouble doing any business at all.

Oracle is intent on quickly turning Sun's hardware operations into a profitable business, and it just might succeed in that goal. But between Sun's weak competitive position and Oracle's shortsightedness, I wouldn't count on seeing the division make much money for Larry Ellison over the long haul.