Today I noticed one of server used for web request profiling stats logging is taking about 2GB per day for logs, which are written in MyISAM table without indexes. So I thought it is great to try how much archive storage engine could help me in this case.

Original Table:

PLAIN TEXT SQL:

1. *************************** 1. row ***************************
2.            Name: requests_061111
3.          Engine: MyISAM
4.         Version: 9
5.      Row_format: Dynamic
6.            Rows: 8048913
7.  Avg_row_length: 252
8.     Data_length: 2030206088
9. Max_data_length: 4294967295
10.    Index_length: 1024
11.       Data_free: 0
…

As you probably know PHP "mysql" extension supported persistent connections but they were disabled in new "mysqli" extension, which is probably one of the reasons some people delay migration to this extension.

The reason behind using persistent connections is of course reducing number of connects which are rather expensive, even though they are much faster with MySQL than with most other databases.

Not only connects are expensive but you also may run into the trouble establishing number of connections you need. The problem is there can be only so many connections active between Host "Apache" and Host "MySQL": Port 3306 as connection in TCP/IP protocol is identified by pair of IP addresses and pair of ports (local port and remote port). Yes if you're establishing thousands of connections per second you normally do not keep it open for long time, but Operation System does. According to TCP/IP protocol Ports can't be recycled …

Sitting here listening to a talk on Tags at MySQL Camp I keep rolling back to the idea that a lot of website problems right now are highly similar.

Lets take the case of application partitioning. If you want to build a large application you need to come up with a way to partition it... how to do this? Do it by user? Do it by time? Do it by some other object?

The problem is dependent on the application.

What I am surprised is that no one has glued together a single solution to these types of problems. I know how to take apart most applications as an engineer in this problem space, but what I don't see is a framework that simplifies this in a common way (idea for Rails?).

The other surprise? Yesterday I made an offhand comment about "image server". AKA creating a farm of machines to just spit out images/ static content. I've solved this problem enough times that I thought that the solution …

Early on Sunday Day #3, I dragged a few willing participants out for a “different photo” based on the umbrellas in each Lobby. It worked out well. Special thanks to Kynan who ran around to other lobby’s to find additional umbrellas. (He is the one holding the white one, and yes, that’s a utility Kilt).

I’ll be uploading more in this series to My Flickr Photos MySQL Camp 01 soon.

You can get a larger copy of image Here.

MySQL Replibeertion was the last scheduled session on Day 2, but not withstanding there was free beer (a lot of), there was a serious side with a Replication Discussion.

One of the first questions by Jeremy was “Are the any big replication users?” to which Sheeri quickly replied “Are you calling me fat again”.

This was a highly interactive session, here are some of the points from the audience.

Some Uses of Replication

• Backup
• Hot standby
• Scaling
• Data Warehousing
  • Slaves are larger then your database
• For no special reason
• Consolidation of multiple sources
• Support for multiple indexes

Issues

• Can break
• replication lag
• bi-directional …

Continuing on from my Day 1 - Memorable Quotes from the MySQL Camp.

“Are there any big replication users” — Jeremy “Are you calling me fat again” — Sheeri

“Only some of us have problems with interruptions.” — Jeremy to Jay

“It really really sucks in production systems.” — Jeremy About Slave management by Master.

“So there are like 12 people here, it must be the CEO’s turn to talk.” — Marten Mickos MySQL CEO

“Kegs and Eggs” — Joel S. Regarding all beer that will still be available at breakfast tomorrow.

“You can fight to the death for it”, Jeremy to his two employees Joel and Justin about who gets to be called employee #1.

“Patches go to employee #1″ — Ronald directed to Joel when a replication patch was coined by Jeremy and …

One think I can see with people using EXPLAIN is trusting it too much, ie assuming if number of rows is reported by EXPLAIN is large query must be inefficient. It may not be the case.

The question is not only about stats which may be wrong and which is why you may want to profile your queries if you have any hesitations in EXPLAIN accuracy.

The other problem however is EXPLAIN does not take LIMIT into account while estimating number of rows. Basically it gives you the estimates of producing whole result set while it may stop much faster in case LIMIT is used.

Take a look at this simple example:

PLAIN TEXT SQL:

1. mysql> EXPLAIN SELECT * FROM rnd  ORDER BY r DESC LIMIT 1 \G
2. *************************** 1. row ***************************
3.            id: 1
4.   select_type: SIMPLE
5.       …

If you have been using solidDB for MySQL Beta 4 and decided to try out solidDB for MySQL Beta 5, you might have seen the following warning message when you try to start up mysqld: ¶

SOLID Database Warning 1022: The database you are using has been
originally created with a different database block size setting
than your current setting.
To suppress this warning edit the solid.ini file to contain
the following parameter setting:
[IndexFile]BlockSize=8192

The reason for this warning message is that the default block size has changed. Beta 4 was using a default block size of 8K and Beta 5 uses a default block size of 16K. ¶

We're setting new XAMPP versions together and would now ask you to take a look at the beta versions of the upcoming XAMPP release.

New in this version of XAMPP: MySQL (5.0.27), PHP (5.2.0), phpMyAdmin (2.9.1) and a Japanese translation.

XAMPP BETA versions are always for testing purposes only. There will be no upgrade packages from and to beta versions. To all testers: Many thanks in advance!!

[b:2c8kx244]Update 18.11.:[/b:2c8kx244] Since yesterday you also find two new versions of XAMPP for Solaris as XAMPP BETA releases, one for Solaris 10 SPARC, and one for x86. These versions are the last step to finalized releases of XAMPP for Solaris. Both versions are compiled for 64-bit systems.

You find the beta version at our special XAMPP BETA download area.

I am thankful for the plentiful feedback from MySQL Camp, on MySQL Winter of Code.

I started today’s session at MySQL Camp by shortly describing our plans, starting by our insight that more coding happens during wintertime than in summer. Through MySQL Winter of Code, we want to encourage contributions to MySQL in all areas of the server. connectors and GUI tools.

There are three requirements for getting a Winter of Code grant:

1. A signed Contributor License Agreement. Highlights:

• You assign and transfer the copyright of your contribution to MySQL. In return you receive back a broad license to re-use and …