paging

Paging is a memory management scheme that eliminates external fragmentation by dividing both virtual address space and physical memory into fixed-size blocks called pages and frames, respectively. A process’s virtual address space is partitioned into equal-sized pages, while physical memory is divided into frames of the same size. A page table maintained by the operating system stores the mapping from page numbers to frame numbers, and a memory management unit translates virtual addresses to physical addresses by consulting this table and a translation lookaside buffer for speed.

During execution, the page number portion of a virtual address is used to index the page table, yielding the corresponding frame number, which combines with the offset to form a physical address. If the required page is not in RAM, a page fault occurs; the operating system must bring the page from secondary storage into a free frame and update the page table, possibly evicting another page using a replacement policy such as LRU or CLOCK.

Benefits of paging include efficient use of memory, no external fragmentation, and strong process isolation. Drawbacks include overhead for maintaining page tables, potential latency from page faults, and the need for fast storage and memory bandwidth to minimize delays.

Variants and related concepts include demand paging, where pages are loaded only when needed; pre-paging, which preloads multiple pages; and inverted page tables, which map frames to pages to reduce table size in large address spaces. Typical page sizes are around 4 KB, with larger “huge” pages used for certain workloads. Paging is a foundational technique in modern operating systems and many computer architectures.