The Sun Rises with Oracle

Why the failure of the IBM deal, and Sun's acquisition by Oracle, is probably good for all three

Friday, May 01, 2009

Print

Comment

Email

Digg

Del.icio.us

Reddit

Twitter

I've written elsewhere that IBM's (now failed) acquisition of Sun was pointless. But this new $7.4 bn Oracle-Sun deal makes sense.

This is an honorable exit for a setting Sun in danger of being gobbled up and swallowed. Oracle won't do that. For it, Sun's hardware is a missing part of the enterprise solutions story.

But the competitor buying a brand to kill it may still be lurking in there. Larry Ellison said nothing of MySQL, Sun's year-old, billion-dollar buy, the open source database that's a thorn in Oracle's side. But as Coke found with Thums Up, it's not easy to kill a great brand. Many new websites use MySQL, not Oracle. My Twitter network tells me that MySQL has been spun off into the open source community under the GPL (General Public License) as Drizzle, so Oracle can't kill it. Still, I'm concerned about Sun's open source projects, including the Open Office that I'm writing this in.

Prasanto K Roy pkr@cybermeda.co.in is president and chief editor at CyberMedia's ICT Publications. With supporting research by Rajneesh De in Dataquest.

Oracle isn't into open source.

When Sun bought MySQL in January 2008, the billion-dollar deal was dwarfed on the same day...by the $8.5 billion Oracle-BEA merger. Oracle wolfs down companies briskly, and digests them well. It's a good buyer. Even if some see more bean counters than tech visionaries in top management.

NEAT FIT?
Ellison called Java the most important software acquisition by Oracle, ever, and spent much media space on Solaris. Yes, those are good fits. (So why did analysts miss this merger? I guess we simply didn't think of it. Once, long ago, I did write about Sun buying Oracle...)

Java is one of the industry's top brands and tech. Oracle's fast-growing Fusion Middleware is built on it. And Solaris is the leading platform for the database that's Oracle's biggest business. With Sun in the bag, Oracle can really tune up its database for Solaris-and vice versa.

From Sun, Oracle gains cloud computing tech. That would pit it well against a SaaS play from SAP or Microsoft; and it doesn't need to buy Salesforce.com now. But Ellison would really like a Solaris-Oracle 11g cloud, rather than a Linux one. Remember 'Unbreakable Linux', Oracle's Red Hat clone? It flopped.

A Twitter friend paraphrases Ellison, “MyCoffee is Java, and MySQL is MyToast”. I hope that stays a joke: that Oracle sees the future in open source, in MySQL-and in an open Java.

THE INDIA STORY
In the Dataquest Top 20 rankings for 2007-08, Oracle was #12 with Rs 5,808 crore from India ops. Sun was at #28, at Rs 1,674 crore. (That's sales plus revenue attributed to global development and support work in India.) Adding the two would have taken the combined entity, in 2007-08, Rs 7,482 crore, or #9 in the DQ Top 20. Well behind HP and IBM, as the #3 software-hardware-solutions player.

On pure India sales, the two are at par: Oracle was at Rs 1,510 crore, and Sun at Rs 1,456 crore (Dataquest estimates), adding up to Rs 2,966 crore in combined sales. That's in the league of SAP (Rs 2,675 cr) and Microsoft (Rs 2,937 cr)... but hey, those are pure-play software vendors. Oracle-Sun should compare with HP or IBM-but it would be dwarfed...three or four times over.

Still, Oracle-Sun would make a strong enterprise solutions play. While the merged duo is much smaller than HP and IBM, this deal could worry the latter two. But later. Given the history of mega-mergers, the tech giants have a year or more to react to Oracle-Sun.

Page(s)   1