6. Concurrent and Scalable Storage Backends

In the previous chapters, we proposed to evade state as much as possible. Our considerations for web server internals are based on a shared-nothing architecture. For programming application logic, we have given the preference to reduce the usage of shared, mutable state to the bare minimum. However, only a handful of web applications can completely disregard any kind of state. Statelessness is viable for validation or computation services, and a few other applications. For virtually all other web applications, some type of persistent state represents an essential part of the overall application. In our architectural model, we have edged away state handling to dedicated backend services.

In this chapter, we consider the impact of concurrency and scalability in storage backends. We illustrate the challenge of guaranteeing consistency in distributed database systems and point to different consistency models. We then outline some internal concepts of distributed database systems and describe how they handle replication and partitioning. Finally, we introduce different types of distributed database systems and assess their features.