In other words, HP could spend nearly $4 billion and get overlapping storage, OS, and hardware products. Sorting out the mess would cost billions more and leave HP even more dependent on Oracle as a partner. One might even say that HP would end up in the pocket of its partner.

On Second Thought…

Scenario two: HP realizes the ugly mess it faces, understands the costs, the headaches, and the customer issues, and it opts not to work with Oracle. This probably wasn't Oracle's first choice of outcome, but it's not a bad one. By buying all of Sun, Oracle finds itself as the only soup-to-nuts player in the data center--from servers, storage, backup, and archive to middleware, databases, and enterprise apps. IBM has a lot of it. With SAP, IBM has it all. Microsoft has it all, too, with Dell and HP as partners, though Microsoft's "it" is more appropriate for midsize customers.

No matter how you slice it, Oracle has carved itself a nice niche that it can defend.

Oracle hasn't made Sun great, but it had made Sun viable, particularly with the addition of its Exa-systems (which aren't SPARC-based, ironically and appropriately). Meanwhile, HP is left to offer platitudes about the software market while its holdings are weak and incomplete, and its financials continue to suffer from its many missteps. HP is a full-service hardware player with holes in its product line (no tablets, no phones), and its software offerings seem arbitrary.

The rift between the two companies combined with HP's management debacles have left HP's market value near its book value, while Oracle's stock continues to rise. Since 2010, Oracle's stock is up about 30% and HP's is down about 50%. Can it all be pinned on this dangerous game the two companies are playing? Almost certainly not, but Oracle has schooled HP and its management on the finer points of big business.

Sorting Through The Crumbs

While Oracle has an impressive win here, one has to ask: Is this sort of bare knuckles business worth it? If stock valuation is the measure, then the answer is no. While Oracle's stock price is up about 30% in about two years, IBM's is up more than 70% over the same period.

IBM's market share in RISC/EPIC systems has been on the rise for a few years now, while both HP's and Oracle's have been declining, just as the overall RISC/EPIC market has been contracting about 15% a year, according to a recent IDC report.

When it comes to selling into the data center, most vendors have been pushing to include servers, storage networking, and services--in Cisco's and EMC's case, through a partnership called VCE. While Oracle/Sun lacks the networking, its offers continue up the stack into middleware, database software, and enterprise applications. Given Oracle's animosity toward SAP, the battle lines become clearer: It's IBM and SAP versus Oracle, with Microsoft taking a healthy chunk of the market with partners HP and Dell.

The question for customers is whether their choice in partners is attractive, and whether they're willing to put all of their eggs in one vendor's basket. In that context, the HP-Oracle suit is little more than a skirmish in a much bigger war. As their non-x86 market shares decline, both Oracle and HP will at some point dump their RISC systems, but the rising competition may not be the traditional foes. Can the likes of Workday and Salesforce.com move enterprises to give up their on-premises back office software?

Fights like the HP-Oracle one and the disregard for customers it displays offer good reasons to look around. But of course, Oracle has that scenario covered too. As of this week, its Fusion line of enterprise applications will all be available in the cloud.

Private clouds are more than a trendy buzzword--they represent Virtualization 2.0. For IT organizations willing to dispense with traditional application hosting models, a plethora of pure cloud software options beckons. Our Understanding Private Cloud Stacks report explains what's available. (Free registration required.)