MySQL, Sun Microsystems' open source database system popular with Web developers, is taking a step closer to the big leagues with the addition of Query Analyzer to its high-end, gold and platinum subscriber editions.
Query Analyzer will become part of MySQL Enterprise Monitor, the management console and production support system that comes with the Enterprise editions of MySQL. The analyzer addition is hoped to push MySQL query execution closer to the performance range of the large commercial systems, which have invested years in query optimization against database operations.
Query Analyzer is meant to address a known weakness of the open source database: Originally designed as a high-speed, read-only database system for serving Web pages, even in its later forms it had little ability to collect query performance data or see performance issues buried in application-level SQL code.
That's changed with the analyzer's ability to look at and identify the most resource-consuming queries on a single database server or set of them. It can tell how frequently queries run, how long they take, and which ones are most prone to create a bottleneck in relation to the value of their results. In the past, MySQL spokesmen have consistently shied away from making any direct comparison between their open source database and large commercial competitors. But that, too, may be changing.
"Oracle DBAs will feel right at home with Query analyzer because Oracle Enterprise Manager (an Oracle add-on management product) does something similar," said Marten Mickos, executive VP of databases at Sun and former CEO of MySQL AB.
"We've been working on it for one-and-a-half to two years. We're doubling the value of Enterprise Monitor, without raising the price," he said.
At Clickability, the San Francisco on-demand, content management supplier to nine NBC stations and several large newspapers, was faced with performance problems buried in queries made to its large MySQL content databases. Jeff Freund, CTO, used pre-product versions of Query Analyzer to see if he could identify bottlenecks and do something about them. It was a daunting task with a billion-row MySQL database that was tracking 5,000 to 10,000 requests per second for content on the Web. Clickability systems serve 400 million Web page views a month across 400 Web sites.
He identified a time-consuming, frequently run query, one that offered to summarize the top sports stories in a given area from various sources. There was nothing wrong with how the query was written other than it was an unbounded query, any number of results was acceptable to it, and it performed both the retrieval and sorting function in the database, Freund learned through Query Analyzer.