In this chapter, you learned how load testing can be used to determine the maximum capacity of message consumers. You learned about message prefetching and how it can improve the performance for consumers that need to deal with large number of messages. We also explained how messaging can facilitate the rolling out of new and unexpected features in a lowly decoupled architecture.

It seems the team in charge of user messages needs some help to deal with old messages piling up in queues; this is what we'll discuss in the next chapter.