What is Virtual Memory?

Virtual memory is a memory management technique that allows the operating system to use part of secondary storage (HDD or SSD) as an extension of main RAM. This creates the illusion that the computer has more RAM than it physically does.

Programs are given virtual addresses — an address space that is independent of the physical RAM installed. The OS maps virtual addresses to actual physical memory locations.

How Virtual Memory Works — Paging

The most common implementation of virtual memory uses paging:

• Both virtual memory and physical RAM are divided into fixed-size blocks called pages (virtual) and frames (physical RAM)
• A page table maintained by the OS maps each virtual page number to its physical frame number in RAM
• Pages that are not currently needed can be swapped out to a reserved area of secondary storage called the swap file (Windows) or swap space (Linux)
• Only the pages currently in use need to be in RAM at any time

Page Fault

A page fault occurs when the CPU tries to access a page that is not currently loaded into RAM:

• The OS pauses the program and locates the required page in the swap file on disk
• If RAM is full, the OS must choose a page to swap out (evict) using a page replacement algorithm (e.g. LRU — Least Recently Used)
• The required page is loaded into the freed frame
• The page table is updated
• The program resumes

Page faults are expensive — accessing secondary storage is much slower than accessing RAM. Frequent page faults slow down a system significantly.

Thrashing

Thrashing occurs when the system spends more time swapping pages in and out of memory than actually executing programs. It happens when:

• RAM is severely insufficient for the number of processes running
• Too many page faults occur simultaneously
• The system constantly swaps pages in and out without making useful progress

Symptoms: computer becomes very slow, disk activity LED is constantly active. Solutions: add more RAM, reduce the number of running processes, use faster secondary storage (SSD).

The Page Table

The page table is a data structure stored in RAM (and managed by the OS) that maps virtual page numbers to physical frame numbers. Each entry also stores:

• Present/absent bit — whether the page is currently in RAM (1) or on disk (0)
• Modified/dirty bit — whether the page has been changed since it was loaded (determines if it must be written back to disk when evicted)
• Referenced bit — whether the page has been recently accessed (used by replacement algorithms)

Translation Lookaside Buffer (TLB)

Every memory access requires a page table lookup (virtual address → physical address), which would be slow if it required a RAM access itself. The TLB is a small, fast cache that stores recent virtual-to-physical address translations:

• TLB hit — translation found in TLB → fast address translation
• TLB miss — translation not found → must access page table in RAM (slower)

Segmentation (Contrast with Paging)

Segmentation is an alternative approach where memory is divided into variable-size segments based on logical units of the program (code segment, data segment, stack segment). Unlike paging (fixed-size), segments have different sizes. Most modern systems use a combination of paging and segmentation.

OS Role in Memory Management

The operating system is responsible for:

• Allocating RAM to processes when they start
• Maintaining the page table for each process
• Handling page faults (loading pages from disk)
• Choosing which pages to evict using replacement algorithms
• Protecting each process's memory space from other processes
• Releasing memory when processes terminate