This is the fourth in a series of articles on profiling MySQL. My past three articles have explained how to measure the work a query causes MySQL to do. In this article I introduce a tool I’ve written to do the work for you and produce a compact, readable report of that work, with all the math already done, and the measurements labelled and grouped for ease of comprehension. With this tool you can understand query performance at a glance.

ZCS 4.0.3 includes fixes for 55 bugs and enhancements. It has significant improvements in calendar behavior and fixes a bug in Trash folder viewing that was troubling many folks. It also includes upgrade speed improvements. This allows the upgrade to take much better advantage of installs that have multiple disks available to MySQL.

Zimbra - Forums - Zimbra Collaboration Suite 4.0.3 Released

ZCS 4.0.3 includes fixes for 55 bugs and enhancements. It has significant improvements in calendar behavior and fixes a bug in Trash folder viewing that was troubling many folks. It also includes upgrade speed improvements. This allows the upgrade to take much better advantage of installs that have multiple disks available to MySQL.

Zimbra - Forums - Zimbra Collaboration Suite 4.0.3 Released

Ok, someone came in #oracle on freenode and wanted a query that would give him a distinct set of rows, where it did not matter if a values is in the first or second column.

Here is a sample table:

CREATE TABLE t (
  a int(11) NOT NULL,
  b int(11) NOT NULL
);

INSERT INTO t (a, b) VALUES (1, 2);
INSERT INTO t (a, b) VALUES (2, 1);
INSERT INTO t (a, b) VALUES (3, 4);
INSERT INTO t (a, b) VALUES (3, 6);
INSERT INTO t (a, b) VALUES (4, 3);
INSERT INTO t (a, b) VALUES (6, 3);

The output should be something like:

(1, 2), (3, 4), (3, 6)
OR
(2, 1), (3, 4), (3, 6)
OR
(6, 3), (2, 1), (3, 4)
etc ..

You get the drift ..

A self join solves the challenge, though we were wondering if any RDBMS out there could do it without a self-join by just using funky analytic features. Any RDBMS specific magic goes (except for implementing a join inside a stored procecure of course). Lets see what …

We have released a new set of bits. The solidDB for MySQL Beta 4 is now out. Like the previous release, it is primarily intended as a bug fix release. Pessimistic concurrency control is now in the product, although it still needs additional testing and is therefore not officially listed as supported in the documentation. ¶

I posted earlier about the MySQL announcement (see http://www.mysql.com/news-and-events/news/article_1171.html )and how I see it affecting Enterprise level users.

However, as someone who uses it for work other than my day job, some at the hobby level and some above that level, the Community edition is also important to me.

The way I see it, the changes are great for people who use the Community edition.

Wait, there will be 2 codebases, and they have not doubled their staff, so won’t there actually be less development? Are they expecting the community to write all the code for the Community edition?

Kind of. I do know that MySQL has been hiring lots of people, but I’m not an employee, so I have no idea of their growth. There will be 2 codebases, but that works out for the best for everyone. Now, there will be 2 …

Interested in being a session speaker at next year's Users Conference? Not put in your proposal yet? Get on it!

In case you didn't read my initial announcement, these are the things I am most looking for in submissions:

• Creativity. Don't be afraid to push the limits with your talk topics. We want to see unique, interesting, and innovative uses and abuses of MySQL!
• Descriptive. The more information you can provide about your session idea, the better, and don't be afraid to go into detail about the pre-requisites you expect session attendees to have in order to get the most out of your session
• Specific. If you want to do a session on stored procedures, great. But, your proposal should be as specific as possible. "Using Stored Procedures to Build a Real-Time Monitoring Solution", "Porting a Dynamic …

Been a while since we've discussed the upcoming MySQL Camp unconference, set for 3 days of rousing good times at the Google Headquarters in Mountain View, from November 10th through 12th. Figured I would post an update, and an appeal for participants to schedule your sessions in the new SessionSchedule.

Currently, there are 24 participants "registered" in the Participants wiki page, but that does not include the ~30-35 Google engineers who likely will attend, according to Leslie Hawthorn. In the next week or so, we will need to start the actual pre-registration process, to enable Leslie to have the security folks know who is coming to the event and have some sort of badge available for you. While the unconference is, of course, free, the registration is required for these security reasons.

I've been throwing …

MySQL today announced that it has become a truly community-driven company. This will sound funny to those of you that think MySQL announced its Enterprise product today. Yes, it did that. But the real news in MySQL's move is its efforts to continue to bolster its already vibrant community.

Savvy users and developers of open source software already know that community is in short supply - there simply is not a large global pool of developers that are anxiously waiting to contribute code to your (or my) project. Therefore, good community developers (and users) are golden, and companies and projects work hard to find them.

In MySQL's case, the company has had to worry …