Demand paging is a technique wherein computer memory is divided into fixed-length blocks. These fixed-length blocks, commonly referred to as pages, can be effectively exchanged or swapped back and forth between primary and secondary storage devices as required. When a page is needed, the location must be immediately determinable. If a page is located in primary memory, it is bound for use in performing the originating object’s objective. That is, memory is relocated and/or attached to permit the completion of the originating program’s functions, even if the page is in a secondary storage device. Generally, control software employs tables to accomplish the task of making pages accessible for processing.
“View Part I of the Managing the Growth and Impact of Virtual Machines and Memory series here“