Queues are a powerful addition to our applications, enabling looser coupling between our components and reducing the load on key elements of the system such as the database. This session covers how queues in general, and excellent open source tool RabbitMQ in particular, can be used to improve the scalability and robustness of applications. We’ll look at designing queue structures and the options around queues, messages, and durability. Creating reliable worker scripts can be tricky but there will be advice on doing so even in systems which require complex or multi-step processing of tasks. We’ll also talk about data hygiene; when to include data with a message, when a worker should fetch data itself, and how to manage the flow of information back to the component that created the message in the first place. With stories and practical examples, this session is recommended for developers and technical architects everywhere.