Memory heirarchy

Memory Management

Memory management refers to the process by which an operating system:

• Allocates memory to processes
• Deallocates memory when no longer needed
• Protects memory from unauthorized access
• Optimizes memory utilization and system performance

Memory Allocation

Memory Allocation is the process of assigning memory space to a program or process.

The OS:

• Keeps track of used and free memory blocks
• Allocates memory when a process starts

Types of Allocation

• Contiguous allocation
• Non-contiguous allocation

Memory Deallocation

Memory Deallocation occurs when a process terminates or releases memory.

Importance:

• Prevents memory wastage
• Makes memory available for other processes
• Avoids memory leaks

Memory Protection

Memory Protection ensures that one process cannot access another process’s memory.

Memory Swapping

Swapping is the process of moving processes between main memory and disk.

• Inactive processes are swapped out to disk
• Active processes are swapped into memory

Virtual Memory

Virtual Memory allows programs to use more memory than physically available.

Key features:

• Uses disk as an extension of RAM
• Only required portions of a program are loaded into memory
• Improves multitasking capability

Memory Mapping

Memory Mapping translates virtual addresses to physical addresses.

Performed using:

• Page tables
• Memory Management Unit (MMU)

Memory Fragmentation

Fragmentation occurs when memory is broken into small unusable pieces.

Types:

• Internal Fragmentation
• External Fragmentation

Memory Paging

Paging divides memory into fixed-size units called pages.

Components:

• Pages (virtual memory)
• Frames (physical memory)
• Page table

Page Table

A Page Table stores the mapping between:

• Virtual page numbers
• Physical frame numbers

Memory Protection

• Memory protection is required to prevent:
  • One process from accessing another process’s memory
  • User processes from modifying operating system memory

Relocation and Limit Registers

• Protection can be provided using:
  • Relocation register - The value of the smallest physical address
  • Limit register - The range of logical addresses

Dispatcher and Context Switch

• When the CPU scheduler selects a process:
  • The dispatcher performs a context switch
• During context switch:
  • Relocation register is loaded
  • Limit register is loaded
• Values correspond to the selected process

Swapping

• Swapping is a memory management technique used by the operating system
• A process must be in memory to be executed
• Swapping allows processes to be temporarily moved out of main memory to disk

Swapping and Priority-Based Scheduling

• A variant of swapping is used in priority-based scheduling algorithms
• If a higher-priority process arrives:

• The OS swaps out a lower-priority process
• Loads and executes the higher-priority process
• After completion:
  • The lower-priority process is swapped back in
  • Execution resumes from where it stopped