Scale Up, Scale Out, and High Availability: Solutions and Combinations

For improving the scalability of your applications you have a lot of potential solutions. Using each individual solution effectively can be complicated enough. But real benefits are gained when you combine different technologies together in different combinations.

This tutorial will demonstrate all of the different basic scale up and scale out solutions and their various benefits and pitfalls. Once the basics of the different combinations have been demonstrated and understood, the tutorial will move on to more complex combinations of solutions that enable you to develop a solution that suits your needs.

The tutorial will cover architectures and topologies for:

• Replication, from basic multiple slave servers to solutions for enhanced performance and throughput by chaining servers together
• Memcached, including how to use it in combination with replication and across multiple servers
• DRBD, both as a HA and a scalability solution
• MySQL Proxy, both as an HA and scalability solution
• Federated and Blackhole tables in scalability environments
• Virtualization and using multiple instances for handling scale out and HA alternatives

The tutorial will also cover using combinations of the above tools to achieve different types and levels of scalability and availability. For example, using MySQL Proxy with replication to provide read scalability, or using replication and Federated tables to support write scalability.

Throughout, full examples and working environments will be demonstrated. These will provide not only samples of the environments and configurations, but also the benefits and the potential problems and how additional solutions can help alleviate or eliminate the problems introduced.