I mentioned a few weeks ago that Jan Kneschke has posted some MySQL user-defined functions -- available here -- for communicating with memcache. I want to outline an example of how to automate cache maintenance by calling these functions from triggers. In a big web application, you have critical high-performance code (which, if you are running memcache, is no doubt already cache-aware), but you probably also have a bunch of little scripts for performing sundry administrative tasks. With triggers in the database, you can keep these scripts the way they are, and even run ad hoc SQL statements, without worrying about leaving inconsistent records in memcache.

This all requires running MySQL 5 with dynamically loaded UDF code -- but of course you don't have to upgrade the whole database farm to MySQL 5, …

Here's an update on the Open Source EAI (Enterprise Application Integration) front as reported a few months back.  Looks like some of the leading open source projects will be merging into one big daddy project.  A combination of ServiceMix (Enterprise Service Bus and SOA Toolkit), Apache Synapse (Web Services Broker) and Celtix (Iona's open source EAI project).  Ioana and LogicBlaze, two commercial companies behind Celtix and ServiceMix, are expected to make the announcement in September. 

One thing that happens fairly rapidly in the open source world is that good projects get momentum quickly.  So I think this is a …

More than at any other time, database professionals are being aggressively challenged by mushrooming volumes of data in corporate business systems. While some industry analysts project data growth at an average rate of 42% a year, that figure is conservative in some installations where growth rates are several hundred percent, with no end in sight.

mysql_pconnect() in PHP... you don't want it. Really.

If you use Apache 1.3, or 2.0 with PreFork, there will be an instance of PHP inside each Apache process. Each of these will have its own cache of persistent connections.

MySQL actually has a very fast connection setup, so it doesn't save time.
With persistent connections, PHP needs to do some cleanup of a connection, to make sure there are no in-progress transactions, or changed server variables. This is a somewhat inexact science. So much so that Georg Richter has actually completely disabled persistent connections in the PHP 5 mysqli extension.

To disable persistent connections, just use mysql_connect() instead.
You can also, very simply, set mysql.allow_persistent = Off in your php.ini file.

Hey all -

I’m busily working away at MySQL on upcoming versions of the server, tools, and more, and would love to get your feedback. Shoot me your “Top 3″ Wish-List for anything that we here at MySQL produce and I’ll make sure it makes its way into our planning process.

Feel free to send me a mail with the heading “My Wish List”. Thanks for helping me and the great engineering staff at MySQL build the absolute best database software!

Robin S
rschumacher@mysql.com

OK, in true journalistic fashion, I've tried to put a sexy, attention-grabbing title on an otherwise not very interesting piece of news. MySQL (surprise!) is in the business of making money. Lots of money. As I've posted before, the company has basically doubled revenues each year for the past several years. Not bad.

I guess if we needed any confirmation that, in fact, MySQL cares about making money, check out this news from eWeek. MySQL is partnering with SCO to bring its database to SCO's OpenServer 6. Ugh. This must be one of those plug your nose and hold your breath sort of moves. I can't imagine that MySQL was happy to be associated with SCO, though the company is right to want to serve its customers. (More on this below.)

So, my …

Please give your opinion: see the quickpoll now running on http://dev.mysql.com/ about the topics you would like to see at the MySQL Users Conference 2006.

Ok, lets start this post with the standard disclaimer about my opinions being my own, and not those of MySQL AB. Why? Because there's controversy afoot.

So what's happening Mike?

Well, dear reader, MySQL AB has gone and added another platform to the list of platforms that we build binaries for. Users of that platform will not only find binaries for their platform, they will also get a trial subscription to MySQL Network to help them use their shiny new MySQL server!

That does not sound so bad, where's the controversy?

Well, depending on who you are, there may be no controversy at all. You see, a press release was made last Friday announcing that the platform in question is SCO Openserver 6.

This has produced a variety of responses, several interesting examples of which can be found …

(insert disclaimer about this being my own views - no that of MySQL AB)

Slashdot | MySQL and SCO Join Forces

Some people seem to think that porting your application to a newer version of an OS, having a trial version of your subscribtion-based support shipping with every copy of that OS and access through that OS vendors reseller channel is a bad thing.

Granted, a lot of people think that certain actions of said OS vendor are just plain retarded. Myself included - it would be much better if they actually focused on products. That being said, there’s more than one OS vendor that does just plain dumb stuff - or, to use the more emotive “evil” word.

Of course, there’s part of the /. crowd that seem to think we must be evil for porting to a SCO platform - but by …

(republished from Slashdot)

So, lets begin this statement with "these are my thoughts, not those that represent MySQL's". First of all I was one of the people who had us stop building on SCO in the first place. For a while now we have not been building for SCO, and had only been providing binaries for customers who had an existing contract with us for those binaries.

The source code for MySQL has always compiled for SCO unixen and since MySQL is open source anyone was free to compile it themselves. We don't ship Amiga binaries either but I can tell you that there is a group out there who keeps MySQL working on that platform as well. So our lack of support for SCO just meant that users were forced to either compile MySQL themselves or find a third party who were distributing the binaries.

Now why should we provide …