Chapter 2. Understanding the PostgreSQL Transaction Log

In the previous chapter, we dealt with various replication concepts. That was meant to be more of a theoretical overview to sharpen your senses for what is to come, and was supposed to introduce the topic in general.

In this chapter, we will move closer to practical solutions, and you will learn how PostgreSQL works internally and what it means for replication. We will see what the so-called transaction log (XLOG) does and how it operates. The XLOG is the very backbone of the PostgreSQL onboard replication machinery. It is essential to understand how this part works in order to fully understand replication. You will learn the following topics in this chapter:

• How PostgreSQL writes data
• Which memory and storage parameters are involved
• How writes can be optimized
• How writes are replicated
• How data consistency can be ensured

In this chapter, you will also learn about replication slots and logical decoding. Once you have completed reading this chapter, you will be ready to understand the next chapter, which will teach you how to safely replicate your first database.