Richard Smith from Sun Labs explains how to code for multi-core systems.

Tim Cook from Sun Labs explains issues and opportunities with locking in InnoDB. - Note that the video must be downloaded to be able to view it!

Databases grow every day more. It is not uncommon to accumulate 100 GB of historical data with an increasing performance problem. MySQL 5.1 with partitions can solve your problems quickly and painlessly in many cases. Using partitions, you can deal with several simple data warehousing cases in a transparent way.

An overview of ZFS, and how to use the functionality of ZFS with MySQL, including performance improvements, to get the best out of ZFS.

mysqlnd, also known as MySQL native driver for PHP, is the new way of connecting from PHP to MySQL. It's a superior substitute to libmysql. mysqlnd is "native" to PHP, in means that it uses the infrastructure provided by PHP. In this session I will show why mysqlnd is better than libmysql, in the PHP world - performance, better resources utilization, quicker bug-fixing cycle and others.

The goal of this session is to introduce users to new standard conforming and engine-neutral implementation of foreign keys, which is being developed for 6.1 version of server. The session will describe capabilities of this new implementation (including limitations for different engines), give a glance at its performance and provide comparison with current InnoDB-specific implementation.

ZFS offers many leading-edge features, including automatic protection against silent data corruption, immense capacity, and vastly simplified administration. But how well does it perform with MySQL? This session explores MySQL performance with ZFS compared to alternative file system implementations. The performance implications of ZFS compression and other features are also examined.

Perl stored procedures provide greater flexibility than the standard SQL stored procedures. For many users, they can provide performance and connectivity which is not available from the built in SQL implementation, where many users have to resort to writing UDFs.

Discussion on solid state disk and how it will change the high performance industry. View pretty graphs and benchmarks that will dazzle your eyes, with numbers sure to get the purest database performance geek excited. DBT2, sysbench, Orion, and other benchmarks will help highlight the advantages and disadvantages of this new and exciting technology.

Although the popular Ruby on Rails framework is known for its ability to quickly build web applications, getting the most favorable performance out of the standard database, MySQL, takes a little bit of effort. This session will cover simple techniques and tricks for optimizing the interaction of the Rails ActiveRecord for MySql.

ScaleDB storage engine delivers high performance transaction processing capability with row level locking and multi-node concurrency control. It enhances performance through innovative indexing technology and delivers low cost computing with a Shared Disk architecture. When applications demand more computing power, additional servers are added to the cluster to scale the system capacity.

You don't have to read source code -- understanding how MySQL works can be achieved through the system data. Learn a wide range of topics in this 3-hour tutorial, and leave armed with tons of knowledge about how MySQL 5.0, 5.1 and 6.0 work. You will take home a healthy understanding of performance tuning, storage engines, replication and many tips and tricks to help you be a better DBA.

Q4M (http://q4m.31tools.com) is a high-performance message queue that works as a pluggable storage engine of MySQL, used by Mixi (Japan's largest SNS provider) and Cybozu Labs (for running Pathtraq, one of the nation's largest web access stats service), etc. The presentation will cover from the design principles of Q4M, to how it is used to modularize, stabilize, and extend the web services.

Connector/J has a wealth of under-promoted performance features. Come to this session to learn about bulk data loading, load balancing, batch rewriting and the "reckless but really fast" options that C/J has.

DTrace is breakthrough tool for on-demand tracing and analysis of any software. As MySQL internals are open, it is possible to extract incredibly valuable information about database work - without performance overhead or even need to restart a server. This talk covers and merges knowledge and practice from both DTrace and MySQL worlds - and explains why D in DTrace means 'dynamic'.

The InnoDB storage engine has been modified to use memcached as a level 2 cache, the InnoDB buffer pool being the level 1. Important reduction of the disk load and performance boost has been observed. This opens a new way of scaling out MySQL.

This session will give an overview of the new set of static DTrace probes introduced in MySQL 6.0 with examples on how to use them in real-life monitoring and performance tuning tasks.

The Falcon Storage Engine is designed to take advantage of multi-core computers. It is reaching GA at a time when computer performance is being increased mostly by adding more cores. Falcon's unique characteristics will be explained and contrasted with InnoDB and other MySQL storage engines.

In a classic case of "eat your own dogfood", MySQL.com is a key player in the beta testing of MySQL software, including Query Analyzer. In this presentation MySQL.com web developer Adam Donnison will take you through how he used Quan to improve database performance on the MySQl.com website by orders of magnitude, and why Quan was needed.

You can run MySQL on a standard server, but there can be benefits to running MySQL within a virtualized environment. You can take advantage of the isolation it provides, or the performance gains possible by running multiple MySQL instances on one large server. This session will show you the benefits and potential issues of running MySQL in a virtualized environment.