Will Sun's MySQL purchase drive down database prices?

Some enterprise database users and industry analysts with whom BetaNews spoke today believe yesterday's announcement of Sun Microsystems' intent to purchase open source database producer MySQL could have an across-the-board effect on the entire database market, impacting IBM, Microsoft, and Oracle.

But based on what we learned today, it doesn't look as though the pricing impact will show up vividly any time soon.

"The pricing on IBM UDB [Universal Database], Microsoft SQL Server, Oracle, and even Sybase databases will come down. But this whole thing will take a long time to play out. It might even take a decade," predicted Raven Zachary, an analyst with the 451 Group, in an interview with BetaNews.

Agreed Charles King, principal analyst at PUND-IT: "Driving down [competitors'] prices is clearly something that Sun wants to do, in taking a database that's popular among some smaller companies and trying to bulletproof it for the enterprise. But this is not exactly an inconsequential effort."

But even if a price war does eventually ensue, customers' database choices are also influenced by a lot of other factors aside from price, according to some close observers. They listed factors such as application support, corporate investment into existing databases, database performance, and in this particular case, Sun's now well-demonstrated commitment to open source.

"If the deal has an impact on [industry] pricing, it still might have only a very indirect effect on which databases people decide to use," said Ted Nichols, VP and CIO at Skyway, a company that's been using the MySQL open source database for the past two years.

Skyway produces a telematic device for automobiles somewhat along the lines of the OnStar system, with a surrounding set of Internet-based services such as theft detection, accident notification, and car diagnostics.

Pricing is very important, Nichols concurred. "But it's really the SAPs, Siebels, and other applications surrounding customers that move them forward to a particular database," he contended.

If customers are ever going to migrate in droves to MySQL, "it'll probably be when somebody develops a real 'killer app' for MySQL," Nichols told BetaNews.

As the 451 Group's Zachary sees it, though, the question of how well an application works with a particular database is one that's mainly important to those who "buy rather than build" -- and companies that purchase packaged applications tend to be newer and smaller.

So, too, is the question of the cost of entry. "These companies are saying, 'We're not running legacy code. We're brand new.' They're not saying, 'We'll keep going with this database as long as it continues to function,'" the analyst told BetaNews.

Large and established enterprises, on the other hand, hold long-term investments in legacy databases - and performance levels could turn out to be the biggest motivator in causing them to switch.

"These companies have made big investments, not just into the software, but into staff and training. They have people working for them who've dedicated their whole careers to a specific database," he remarked.

"But if Sun starts taking on the big databases by optimizing MySQL, that could cause a real shift, especially if Sun can legitimate performance through benchmarking. Nothing much has been done yet about benchmarking open source databases against the IBMs and Oracles."

According to King, though, Sun might find it very tough to raise MySQL's performance to those levels. "If somebody new is going to move in to enterprise databases, somebody else is going to have to move out. Companies have so much strategic investment into their databases that it would take a pretty radical drop in price to inspire them to switch," BetaNews was told.

"And as good as MySQL is as a database, there are probably some development issues ahead. Sun might get some initial success among its current customer base -- but still, they have a lot of work ahead of them."

Yet Sun's acquisition of MySQL -- along with its other open source activities over the past couple of years -- positions the company well with open source advocates, observers pointed out.

"With the MySQL buyout, Sun is making a very visible statement about open source. They've also paid for a user base. How well MySQL does depends less on what Sun does with the product and more on how well Sun exploits its relationship with the existing community," according to Zachary.

Nichols, for one, has no plans of dropping his use of the MySQL database now that the open source vendor belongs to Sun.

But meanwhile, he does not intend to buy a license from Sun that would entitle him to database support, just as he had chosen not to have purchased one from MySQL before the acquisition.

Instead, he'll continue on with a contract his company holds with OpenLogic for support across all of the start-up's open source middleware and applications: MySQL, JBoss, TomCat, and MediaWiki.

Nichols acknowledges that he gets a better overall deal on pricing through this blanket approach.

But, he says, his decision is also spawned by the difficulties inherent in holding licenses with multiple vendors, along with his desire to avoid "vendor finger-pointing" if a software problem crops us somewhere along the way.