If IBM Buys Sun, What Does That Mean For MySQL?

Before I go on, let me be clear -- this deal is still rumored and while reports suggest it could happen in days, it could fall apart. Now on for the speculation.

Of course the repercussions of such a megadeal transcend way beyond one particular component of Sun's arsenal, which includes a contracting server business that is nonetheless well regarded technically, microprocessors, software and a deep bench of R&D. And of course there's perhaps Sun's most visible asset -- Java.

Some question what IBM would do with MySQL, considering Big Blue has the means to clone it with an express-type version of its DB2 database. Dana Gardner, principal analyst with Interarbor Solutions LLC, a Gilford, N.H. consultancy, is among those who sees no sense in IBM acquiring Sun, as he posted in his blog yesterday.

"There's no reason for IBM to take open source any further than it already has, given that it still commercially has a very good business in DB2," Gardner said in a subsequent interview. "If they want to further their open source database strategy, they can accomplish that without buying Sun; they could buy Ingres or spin off an open source version of DB2."

Forrester analyst Noel Yuhanna disagrees. MySQL has cache with customers that no other open source database has achieved, Yuhanna says. Moreover it has more revenues than any other open source database -- $400 million for the most recent year, he estimates, though he points out that's miniscule compared to the overall $16 billion market for database software.

MySQL, Yuhanna told me, would give IBM an opportunity to offer an alternative database to customers looking to move to an open source database.

"In these economic conditions today, companies are looking at open source databases more aggressively because they want to lower the cost, and my SQL is very mature in terms of technology," Yuhanna said. "They have that niche of becoming a more scalable, high-performance database."

Many customers see DB2 as a mainframe-class database, he added, and IBM has failed to make strong gains with that database on Windows and Linux. "If you look at the mix, it would be really complimentary for IBM," he said. Analyst Curt Monash, of Monash Research agrees. "There's little reason to think IBM would orphan MySQL or any other DBMS product," Monash wrote in a blog posting.

One of the key areas Sun has failed in growing the MySQL business is its lack of migration services and tools from higher end databases, Yuhanna added. Many Forrester clients have indicated that Sun hasn't improved MySQL with performance and scalability. "They fear if they don't provide a very good support for high end, we may just move away from MySQL," he said.

A resurgent MySQL would perhaps most be competitive with Microsoft's SQL Server and with its recent acquisition of DATAllegro, is gaining more credibility for high end implementations including business intelligence and data warehousing.

Yet Yuhanna says .NET customers prefer SQL Server for obvious reasons, especially now as Microsoft is providing even tighter integration with its framework and tooling. "That's where IBM would have to attack, try to provide an integration point with Java and MySQL to make it more appealing to customers."

Of course, we'll see if this all happens. If you have any thoughts, drop me a line at [email protected].

Posted by Jeffrey Schwartz on 03/19/2009