<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.3.3" -->
<rss version="0.92">
<channel>
	<title>QOT</title>
	<link>http://ritmark.com</link>
	<description>MySQL DBA's intelligent assistant</description>
	<lastBuildDate>Sat, 15 May 2010 19:50:55 +0000</lastBuildDate>
	<docs>http://backend.userland.com/rss092</docs>
	<language>en</language>
	
	<item>
		<title>Formal verification of PBXT locking code, my experience with verification tools</title>
		<description>In the last few months we've got couple of bugs with the PBXT read/write locking code. One of the problems was directly in the lock/grant algorithm. It was strange to find such kind of error after quite long time of various testing and real-life usage. From the point of view ...</description>
		<link>http://ritmark.com/?p=28</link>
			</item>
	<item>
		<title>Last week in QOT: Index/Predicate Selectivity Analysis</title>
		<description>Last weekend I've reached a milestone in my work on the statistical index analysis for QOT. The code is available on Launchpad.

An example is worth a thousand words, so I start with one (once again I'm using employees-db):

$ ./qot --info --host=192.168.200.1 --analyse=possible --schema=employees --query="select * from dept_emp where (dept_no = ...</description>
		<link>http://ritmark.com/?p=27</link>
			</item>
	<item>
		<title>Selectivity threshold for a non-covering index</title>
		<description>Assume you have a table with about 300 000 rows, and an indexed column 'col1' with only 9 distinct values. Now you got a query like 'select * from t1 where col1 = const'. The questions are

- when the index is faster to full table scan and vice versa?
- does ...</description>
		<link>http://ritmark.com/?p=25</link>
			</item>
	<item>
		<title>Redundant &#8220;Using where&#8221;</title>
		<description>Working on server-based index analysis I once again recalled a topic I wanted to write about for many times. The topic is redundant "Using where" in query plans. Suppose we have a table like this:

mysql&#62; show create table t1 \G
*************************** 1. row ***************************
Table: t1
Create Table: CREATE TABLE `t1` (
`a` int(11) ...</description>
		<link>http://ritmark.com/?p=24</link>
			</item>
	<item>
		<title>Test framework migration finished, started work on server connections</title>
		<description>Finally QOT development has reached an important milestone - all the tests were migrated to the new testing framework. The latest changes can be found at Launchpad - lp:qot. This switches the green light to the development of new features.

The first thing to be implemented is data schema reading from ...</description>
		<link>http://ritmark.com/?p=22</link>
			</item>
	<item>
		<title>Another Week in QOT</title>
		<description>Last week (or rather last weekend) was quite productive. I was able to make a good progress migrating to the new mysql-test based testing framework. Reviewing the tests I was able to find and fix some bugs. For example there was a bug that ignored some columns while detecting index ...</description>
		<link>http://ritmark.com/?p=21</link>
			</item>
	<item>
		<title>Last week in QOT: bugfixes, mysql-test</title>
		<description> With this post I want to start the practice of writing short weekly summaries of QOT activities.

This week I was able to work on some bug reports: #12 #13 #14 #15. Some trivial crashes, some of which were already fixed before. As I wrote in my previous post, I have ...</description>
		<link>http://ritmark.com/?p=20</link>
			</item>
	<item>
		<title>Changes in the Development Model</title>
		<description>General

Being quite busy with my full-time job at PrimeBase Technologies and quite limited in time for QOT development I was looking for ways to optimize the process around QOT.

I came to a conclusion that instead of preparing and releasing milestone binary releases it will be more optimal to declare the ...</description>
		<link>http://ritmark.com/?p=19</link>
			</item>
	<item>
		<title>administrative: this should have happened sooner or later and it finally happened&#8230;</title>
		<description>I had to disable write access for anonymous mantis account because of the spam in comments. Will enable it one I add some kind of protection. For now - please register to report a bug. Thanks. </description>
		<link>http://ritmark.com/?p=18</link>
			</item>
	<item>
		<title>QOT moved to Launchpad</title>
		<description>I moved all the QOT code to Launchpad. This includes code for the tool, tests and a small experimental utility qot-predictor, which based on query analysis tries to predict which queries are likely to be executed next. Project URL for branching is lp:qot. If you want to push your own ...</description>
		<link>http://ritmark.com/?p=17</link>
			</item>
</channel>
</rss>
