Finally I was able to run PBXT 1.0.11 pre-GA in tpcc-like workload, apparently there was bug with did not allow me to get the result earlier, and I am happy to see that PBXT team managed it.

For initial runs I took tpcc 100 warehouses ( about 10GB of data) which fully fits into memory (32 GB on server),
and compared 1 and 16 users in MySQL-5.1.46/PBXT and Percona Server / XtraDB - 5.1.45-rel10.2. As workload is totally memory based it will show how PBXT scales in CPU-bond cases on 16 cores systems.

As storage system it was Intel SSD X25-M card.

While full results and config are on Wiki:
http://www.percona.com/docs/wiki/benchmark:pbxt:tpcc:start

there are graphs for 1 user:

When I have discovered MySQL User-defined variables at the first time, I didn’t pay much attention to them and didn’t realize their strength. At first, one can think that they are just one of the many unnoticed MySQL features. Well, I was wrong. Here, I will try to summarize their strength, though I think there is […]

Dear MySQL Users,

We’re happy to announce the second Release Candidate (RC) of MySQL Workbench.
Version 5.2.22 includes fixes for 76 bugs.

In case you didn’t know – MySQL Workbench is a modular product and is fully pluggable and scriptable. If you want to write a plug-in for Workbench 5.2, then stay tuned as we will be updating the documentation in the next week with all the details and examples you need on how to write your own plug-ins.

Plug-ins are great for extending or adding to workbench or for automating repetitive tasks. You can do things like adding to menus, exporting your Workbench models custom tailored for your coding environment, adding your own forms, and more. Python is our preferred scripting language but for simple scripts the Lua language can be used as well.

The first plug-in we added is called MySQL Doc Library. It can be used to read the MySQL Documentation offline and features …

The mysqlnd plugin API is a well hidden gem of mysqlnd. Mysqlnd plugins operate on a layer between PHP applications and the MySQL server. This is comparable to MySQL Proxy. MySQL Proxy operates on a layer between any MySQL client application, for example, a PHP application and, the MySQL server. Plugins can take over classical MySQL Proxy tasks such as Load Balancing, Monitoring and Performance optimizations. But due to the different architecture and location mysqlnd plugins do not share some common MySQL Proxy annoyances: no single point of failure, no dedicated proxy server to deploy, no new programming language to learn (Lua).

Slides from the IPC Spring conference

This blog posting and today’s presentation at the IPC Spring 2010, a …

The mysqlnd plugin API is a well hidden gem of mysqlnd. Mysqlnd plugins operate on a layer between PHP applications and the MySQL server. This is comparable to MySQL Proxy. MySQL Proxy operates on a layer between any MySQL client application, for example, a PHP application and, the MySQL server. Plugins can take over classical MySQL Proxy tasks such as Load Balancing, Monitoring and Performance optimizations. But due to the different architecture and location mysqlnd plugins do not share some common MySQL Proxy annoyances: no single point of failure, no dedicated proxy server to deploy, no new programming language to learn (Lua).

Slides from the IPC Spring conference

This blog posting and today’s presentation at the IPC Spring 2010, a …

In Percona Server security fix releases I mentioned patches for MySQL 4.0 and 4.1.

I am happy to announce that GoDaddy.com released patches for MySQL 4.0 and MySQL 4.1 under GPL license and you can get them from our Launchpad:

for 4.0: lp:~percona-dev/percona-patches/4.0.30 ( or https://launchpad.net/~percona-dev/percona-patches/4.0.30 )
for 4.1: lp:~percona-dev/percona-patches/4.1.24 ( or https://launchpad.net/~percona-dev/percona-patches/4.1.24)

Fixed bugs:

• Bug#53371, …

As mentioned in parts 1 and 2, having many disk seeks are bad (they slow down performance). Fractal tree data structures minimize disk seeks on ad-hoc insertions, whereas B-trees practically guarantee that disk seeks are performed on ad-hoc insertions. As a result, fractal tree data structures can insert data up to two orders of magnitude faster than B-Trees can.

In this post, let’s examine deletions, and get an intuitive understanding for why fractal-tree data structures exhibit the same two orders of magnitude faster deletions than B-trees. In MySQL 5.1, this advantage is really eye-popping for TokuDB v. InnoDB, because InnoDB does not use its insert buffer for deletions. I understand there is a delete buffer in 5.5, which I …

This Thursday (June 3rd, 14:00 UTC), Reggie Burnett, head of Connector/NET development, will present the New Features in Connector/NET 6.3.

For MySQL University sessions, point your browser to this page. You need a browser with a working Flash plugin. You may register for a Dimdim account, but you don't have to. (Dimdim is the conferencing system we're using for MySQL University sessions. It provides integrated voice streaming, chat, whiteboard, session recording, and more.)

MySQL University is a free educational online program for engineers/developers. MySQL University sessions are open to anyone. All sessions (slides & audio) are recorded; the links will be on …

When analyzing how good or bad response time is it is not handy to look at the averages, min or max times - something what is easily computed using built in aggregate functions. We most likely would like to see some percentile numbers - 95 percentile or 99 percentile. The problem is computing these in SQL is ugly and resource consuming. There is however a simple way to get similar data, looking at it from the different point of view.

When we're speaking about application we may not always care about exact value of response time but rather we want to see response time to be within certain range. For example if we define page feels good if response time is below 50ms it is not as important if response was 40ms or 44ms - it is much more important how frequently this goal was reached.

In fact I prefer to define 2 performance level. One what users would consider good performance, another what is acceptable, and basically the third level …

A couple of weeks ago I gave a presentation at the Apache Lucene Eurocon in Prague. It was a good conference focused on Lucene/Solr open source search technology and sponsored by Lucid Imagination.  

I've posted the bulk of the presentation below.  (I omitted a couple of slides that were MySQL specific.) Even though it was a technical conference, I got positive feedback from the attendees and organizers that the information was useful in helping folks think about where to focus their efforts.  

The slides have been posted to Box.net and are shown using their new "embedded preview" feature which is pretty cool. …