As you will be aware, the kernel tries, for optimal performance, to keep the working set of memory pages as high up as possible in the memory pyramid (or hierarchy).
The so-called memory pyramid (or memory hierarchy) on a system consists of (in order, from smallest size but fastest speed to largest size but slowest): CPU registers, CPU caches (LI, L2, L3, ...), RAM, and swap (raw disk/flash/SSD partition). In our following discussion, we ignore CPU registers as their size is minuscule.
So, the processor uses its hardware caches (L1, L2, and so on) to hold the working set of pages. But of course, CPU cache memory is very limited, thus it will soon run out, causing the memory to spill over into the next hierarchical level – RAM. On modern systems, even many embedded ones, there's quite a bit of RAM; still, if and when the OS does run low on RAM, it spills over the memory pages that can no longer fit in RAM...
