PBOs "pipelined"?

My understanding of GL is that of a beginner, and a rusty one at that. I’m reading the 5th edition of the superbible at the moment, and in a paragraph on PBOs it says:

Why bother with these pixel buffer objects anyway? After all, you can get pixels to, from, and around the GPU without them. For starters, any calls that read or write to PBOs or any buffer object are pipelined.

What exactly does “pipelined” mean here? What are the practical consequences (in low level terms) of being pipelined as opposed to not being pipelined?

(If anyone can help explain this term, then feel free to assume in your answer that I do understand a bit about how the graphics driver queues up and sends commands to the GPU, and about how there is a very physical separation of the CPU and GPU, and that they have a very client-server like relationship).

What exactly does “pipelined” mean here?

I have no idea why he called it “pipelined”. The proper term is “asynchronous”. The OpenGL Wiki page on this matter can explain in greater detail.

It means that updates to them shouldn’t block the CPU or GPU. Multiple blocks of memory actually exist behind buffer objects to make them pipeline well.