For example, 25 years ago, techniques for moving data through memory hierarchies were important.

The system interface and memory hierarchy was also significantly reworked.

The lower levels of the memory hierarchy tend to be slower, but larger.

One of the main ways to increase system performance is minimising how far down the memory hierarchy one has to go to manipulate data.

Predicting where in the memory hierarchy the data resides is difficult.

Then the memory hierarchy will be assessed during code refactoring.

The state of the memory hierarchy is checkpointed by disabling stores.

Processor registers are normally at the top of the memory hierarchy, and provide the fastest way to access data.

This automatically takes advantage of two or more levels of memory hierarchy, if available.

The reason we have a memory hierarchy in the first place is because the price of faster memory is too much to scale up.