|
Mark Matthews on MySQL Query Analyzer What It Does to Make Life Easier for the DBA
Mark Matthews is one of the architects of the MySQL Query Analyzer and, therefore, a logical choice as in interview subject on that aspect of the MySQL solution. At the outset, Matthews explains the thinking behind MySQL Query Analyzer, which is that its development and use would make MySQL Performance Tuning for DBAs and developers significantly less of a mystical operation requiring esoteric knowledge. He added that MySQL Proxy is a tool that also works to simplify the use of such applications as MySQL Enterprise Monitor, making it possible to see at any given moment which queries were running where across an entire farm of MySQL servers. "MySQL Query Analyzer uses Proxy to capture aggregate metadata about queries flowing through it and send that information to a repository for analysis," Matthews continued. With MySQL Proxy it is possible to develop a perspective from inside a query stream, displaying information critical to a DBA's purposes, information that MySQL alone cannot provide. The MySQL Query Analyzer captures every query, times them and reports them in aggregate, back up to the Service Manager repository, making these data available for analysis across all one's servers or simply across groups of servers, sorted by query type. "You can search through them using regular expressions, you can see the minimum, maximum, average and distributions of execution times, execution counts, row and update counts, result set sizes, and actual amount of data transferred and you can even bring up EXPLAIN plans of queries that have been captured without leaving the application," Matthews explained. Even more useful, is the ability of the MySQL Proxy to be transparent to the application other than sending the queries to a different port, which means the queries get looked at by the Proxy (but are not changed in any way) and forwarded on to MySQL, according to Matthews. The reporting itself is done with data that winds up in the Service Manager. DBAs will welcome MySQL Query Analyzer, Matthews is confident, because it does the work of a panoply of earlier tools that each provided partial information where Query Analyzer does it all. Matthews pointed out how MySQL Query Analyzer was able to detect a query that was part of a newly developed application and that was being issued, superfluously, at the rate of 400,000 times per minute. This query would not have shown up on the Slow Query Log, and the resource-consuming overhead would not have been discovered. Developers, Matthews noted, find MySQL Query Analyzer especially helpful in its ability to reveal what queries an application is issuing and in what number. Another useful feature of MySQL Query Analyzer cited by Matthews is its ability to identify queries that are candidates for caching. DBAs can easily determine what the resource cost of any aspect of an application is. Matthews said this feature is removing much of the drudgery that used to be common in a developer's day. As to the future, Matthews announced that users to graph CPU, I/O, or database activity, highlight it and go to a page where you have all the query analysis results that are filtered based on the time window you specified. Asked if it might become possible to use MySQL Query Analyzer without needing Proxy middleware Matthews responded by saying, "We don’t want MySQL Proxy to be the only source of the metrics we are collecting." Users eager for a taste of MySQL Query Analyzer can access it as part of the MySQL Enterprise Trial for 30 days without even being a MySQL Enterprise customer. More Information
Read the complete text of the Mark Matthews interview here.
[...read more...]
Keywords:
From the latest issue:
|
|
|
|
|
|||||||||||||||||||||||||||||||||||||||||||||
Print Friendly
Email this article
reveal what queries an application is issuing and in what number
Mark Matthews on MySQL Query Analyzer
Copyright © 1998-2009 System News, Inc.
