Asynchronous IO allows us to send a request without waiting for its result, we'll get notified somehow later when we receive the response. This enables our programs to be more concurrent and scale better.
In the previous chapter, we used threads in order to avoid blocking other clients while we wait for a response. Using threads has a cost, though, besides the fact that threads require more memory, they also impose a performance cost because they require a context switch when the code goes from one thread to the other.