PHPBuilder - MySQL, today's contender Page 2

RSS Twitter

MySQL, today's contender - Page 2

by: Ian Gilfillan
May 15, 2007

However, there is something positive planned for MySQL 5.2 - foreign keys. Yes, you read right, foreign keys! They have been supported by the InnoDB storage engine since the days of MySQL 3.23, and the syntax has been supported by the other storage engines for a while, so most developers have hardly noticed their omission. However, from version 5.2 all storage engines, including the MyISAM engine will fully support them. MyISAM was originally designed as a light, fast, storage engine. With all the new features added since its humble beginnings, it will be interesting to see how its performance suffers.
What's missing?
The big difference between the new features being added in upcoming versions of MySQL, and those that were still being added two or three years ago, is how less likely an experienced DBA, who is a newcomer to MySQL, will be surprised that the feature is not there. Before, absolutely vital features such as views, triggers, subqueries and stored procedures were missing. All of these necessities are now there, and the new features are more from the realm of nice-to-haves for most developers. If you are ardently disagreeing at this point, claiming the absolute necessity of something like the event scheduler, look around, you are the exception that proves the rule.
There is little missing in terms of features now. Instead, MySQL's status as a contender is now more dependant on 3rd-party support, and integration into enterprise applications, where it still lags significantly. However, the doors are now open, and there is little technically stopping more widespread adoption. The murky realms of marketing and deal making will determine MySQL's future in that arena.
Recognition as a contender
Perhaps the greatest recognition that MySQL has arrived (wherever exactly that would be) has come from Oracle. Oracle, still broadly perceived as the leading database vendor, have purchased Innobase OY, who are responsible for MySQL's most advanced storage engine, InnoDB. Shortly after, they purchased Sleepycat, who are responsible for the BDB storage engine. Both these storage engines add much-needed functionality, without which MySQL again reverts to toy database status. Moreover, most indicative of all, Oracle made an offer to purchase MySQL, which was turned down.
However, MySQL has responded in the way one would expect. They've looked at securing the expertise to build their own transactional storage engine, and have done so by buying Jim Starkey's Netfrastructure, and securing the services of Jim Starkey. Starkey is highly-regarded as the father of Interbase, which later forked into Firebird. MySQL have also secured a multi-year deal with Oracle renewing their InnoDB licensing, which ensures stability for a while.
At the same time, Starkey and others have clearly been hard at work. A new storage engine, Falcon, should be ready for beta testing soon. Falcon will be a fully-featured transactional MySQL storage engine, based on the mature Netfrastructure engine, which has been in use of over 4-years, and has now been integrated into MySQL. As Starkey made clear in a presentation at the 2006 MySQL Users Conference in Santa Clara in late April, Falcon is not an InnoDB clone, a Firebird clone, a standalone DBMS or Netfrastructure. It's a fully-featured storage engine that's likely to become the de facto standard for MySQL.
MySQL today is in a healthy position. The company supporting it is making good revenue, its technical development has come along well to the point where it is sufficient for the vast majority of the market, and there is great expertise continuing to develop it. It is positioning itself so that even if Oracle's moves are threatening, it will come through without much disruption. The biggest risk for its future is probably if MySQL AB accepts a buyout offer from Oracle, and as MySQL continues to make inroads into Oracles database market, Oracle will can respond by throwing money at MySQL to make the problem go away (which they've already tried), as well as positioning themselves as more than just a database company, which they are doing relatively successfully.
This article originally appeared on

« Previous Page

Comment and Contribute

Your comment has been submitted and is pending approval.

Ian Gilfillan



(Maximum characters: 1200). You have characters left.