Here is a bit of code I worked up for us a recently for Drizzle:

drizzle> DELIMITER |
Note that there is no semicolon after the '|' symbol, which we will use as the delimiter for our purposes. You have to choose a delimiter that does not appear in your procedure, and it can be more than one character.

drizzle> CREATE PROCEDURE perl_hello (param1 string)
-> return "Hello " . $_[0] . "!"
-> |
Query OK, 0 rows affected (0.05 sec)

drizzle> CALL perl_hello('Brian');
-> |
Query OK, 1 row affected (0.00 sec)

drizzle> DELIMITER ;
drizzle> SELECT @perl\G
*************************** 1. row ***************************
@perl: Hello Brian!
1 row in set (0.00 sec)


Stored Procedures!?!

In an actual language!?!

About a week ago I was talking to a CTO for a …

I thought I would put out a quick note about this right now, I am going to be leaving Sun next week. I do have some plans which will start to materialize over the next few weeks and months so stick around for that. Even though I am leaving the company, I will still be lurking and working on all think database and performance related. This should not have any bearing on BigDBAHead.com or Waffle Grid, hopefully this will only mean I will post more and spend more time doing crazy and cool things!

I wanted to thank everyone at MySQL who I worked with and learned from over the past few years.

As someone may have noticed, I recently wrote a trilogy on how to dive into the MySQL Cluster source code. Unfortunately my overtures towards the MySQL Cluster source code ended up being only a look-but-don't-touch affair, as I failed to actually get to touch her internals with my text editor. Even so, in this post I'd like to tell about the background to my love affair with this beauty, by relating to some benchmarks I've been working on together with my customers.

Oh, and I'd like to apologize already, that I cannot mention where these benchmarks were done, what the schema looked like and the exact numbers. If you want that kind of real benchmarks, you should read Mikael's blog, or watch the slides from this …

As many of you who follow the drizzle-discuss mailing list know, for the last three months, Drizzle developers have been hot (and many times cold) on the tail of a performance problem that we were seeing when comparing Drizzle with MySQL (any version of MySQL, not just 5.4).

Briefly, on certain machines, we were seeing Drizzle performing at approximately 50% of MySQL, with throughput measured in transactions per second on both a readonly and readwrite workload with Sysbench. The frustrating part of the results was that on other machines, even other machines with virtually identical architectures, compilers, and operating system, we were seeing Drizzle outperforming MySQL by around 20-30%.

So, the Sun Drizzle team, friends in the Sun PAE (performance applications engineering?) team and various contributors set out to …

MySQL clients uses COM_\* commands to communicate with the MySQL server. MySQL show status breaks these commands into many categories and displays how many times each of these commands were executed. It, however, does not display the time taken to processes those commands as well as how many times the command execution resulted in an error. Enter Dtrace

I wrote a small DTrace script to figure out what commands(COM_\*) are being executed on the server and summarize them. If you have used truss -c on Solaris before, you must liked its concise summary; I have tried to present the output in a similar fashion.

For Sysbench read-write (10 queries per transaction) test with 1 thread, each executing 1 transaction, you see

# ./cmdtruss
Sampling... Hit Ctrl-C to end.
\^C
Command                  seconds     calls  errors
Query                      0.042         2        
Quit                       0.000         2        
CloseStmt …

MySQL clients uses COM_\* commands to communicate with the MySQL server. MySQL show status breaks these commands into many categories and displays how many times each of these commands were executed. It, however, does not display the time taken to processes those commands as well as how many times the command execution resulted in an error. Enter Dtrace

I wrote a small DTrace script to figure out what commands(COM_\*) are being executed on the server and summarize them. If you have used truss -c on Solaris before, you must liked its concise summary; I have tried to present the output in a similar fashion.

For Sysbench read-write (10 queries per transaction) test with 1 thread, each executing 1 transaction, you see

# ./cmdtruss
Sampling... Hit Ctrl-C to end.
\^C
Command                  seconds     calls  errors
Query                      0.042         2        
Quit                       0.000         2        
CloseStmt …

XtraBackup is an Open Source online (non-blockable) backup solution for the InnoDB and XtraDB storage engines. It works with both MySQL 5.0 and 5.1 (and possibly 5.4 as well) and is distributed under the GPLv2.

Some weeks ago Vadim announced the availability of xtrabackup-0.7, stating that they consider it stable enough now to label this version a "Release Candidate". I've been maintaining RPM packages of xtrabackup on the fine openSUSE Build Service for quite some time now, RPMs of 0.7 for a number of distributions are now …

The German Oracle User Association (DOAG e.V.) has published a statement (in German) about the acquisition of Sun/MySQL by Oracle and its impact for Oracle users. You can find the statement here.

Oh and btw, I'll give a session about "PHP5 & Oracle" at the local Oracle usergroups in Frankfurt on June, 23rd and Hamburg on Sep 14th. The main goal is to promote the usage of PHP5 in Oracle environments (and how you can leverage PHP's potential in Enterprise environments) as there are good Oracle database connectors for PHP5 available. See you there!

Thanks for the overwhelming response to SQLyog 8.1 release. We had highlighted the speed improvements in Data Synchronization in that post. As promised here are the benchmarking results. In this post, we will compare SQLyog 8.1’s data sync speeds against previous version of SQLyog (v 8.05) and Maatkit (v 3519).

We have done similar comparisions in the past. By comparing against Maatkit, we are not trying to imply that SQLyog is better than Maatkit or something similar. Both these tools target a different set of audiences. Maatkit has got some incredible tools that we use internally for a variety of tasks. Maatkit is command line, free and open source. SQLyog’s Data …

In addition to being practical, …