PromleeBlog

posting thumbnail
What Is Virtual Memory? — 10 Minutes of Operating Systems, Day 11

📅
Before We Begin
What Is Virtual Memory?
Why Do We Need Virtual Memory?
Demand Paging
How It Works
Example Flow
Page Replacement
FIFO
LRU
Optimal
Copy-on-Write
How It Works
Memory Compression
Kernel Memory Allocation
Summary
References

Before We Begin 🔗

The programs we use need much more memory than we might expect. Physical RAM, however, is limited. The operating system has a clever technique for this situation:
virtual memory
.
In this post, we will work through virtual memory, how pages are loaded and replaced, and the often confusing concept of copy-on-write, explaining each step simply enough for a child to follow.

What Is Virtual Memory? 🔗

Virtual memory
is an operating system technique that creates an address space larger than physical memory. A program can behave as though it has an enormous amount of memory, while only part of its data is actually in memory and the rest is stored on disk.
Imagine living in a small room, keeping a separate storage unit, and bringing out only the belongings you need.

Why Do We Need Virtual Memory? 🔗

Demand Paging 🔗

Rather than loading every page into memory at the start, virtual memory loads pages
only when they are needed
. This is called
demand paging
.
📦 It is like taking only the items you need out of storage.

How It Works 🔗

  1. The program starts → only the first few necessary pages are loaded into memory.
  2. Another page is needed during execution → a
    page fault
    occurs.
  3. That page is retrieved from disk → it is loaded into memory.
  4. Execution resumes.

Example Flow 🔗

  1. A process has 10 pages, but only 3 are loaded into memory.
  2. When another page is needed, the operating system detects a page fault.
  3. The required page is copied from disk to RAM.
Demand paging lets us make more effective use of memory.

Page Replacement 🔗

The problem arises when memory is full. What should we do if a new page needs to be loaded but there is no room?
🎒 If your bag has too many books, you take out one you are not reading and put in a new one.
The operating system uses a
page replacement algorithm
to decide which page to remove.
Common algorithms include:

FIFO 🔗

Remove the page that was loaded first.
페이지 프레임: 3개
요청: A, B, C, A, D

→ 적재 순서: A, B, C
→ A는 재사용됨
→ D 요청 시, A가 가장 먼저 들어왔으므로 A 제거
→ 페이지 교체: A → D

LRU 🔗

Remove the page that has not been used for the longest time.
페이지 프레임: 3개
요청: A, B, C, A, D

→ A, B, C 적재
→ A 재사용됨 → A는 최근 사용됨
→ D 요청 → B가 가장 오래 사용되지 않음 → B 제거
→ 페이지 교체: B → D

Optimal 🔗

Remove the page that will be used farthest in the future. This is difficult to use in practice because it requires knowing the future.
페이지 프레임: 3개
요청: A, B, C, A, D, B

→ A, B, C 적재
→ A 재사용됨
→ D 요청 → C는 앞으로 사용되지 않음 → C 제거
→ B 요청 → A는 안 쓰고 D는 나중에 → D 제거 (미래 기준)

Copy-on-Write 🔗

When two processes use the same data, they initially
share
the same pages. A page is
copied only when one process attempts to modify it
.
📄 Think of two people reading the same document, with a separate copy made only when one of them wants to edit it.

How It Works 🔗

  1. A parent process creates a child process, and they share memory.
  2. Both can read the memory.
  3. If either tries to modify data →
    only the affected page is copied and separated
    .
This technique is very commonly used for efficient memory management after
fork()
.

Memory Compression 🔗

Instead of swapping, some operating systems
compress unused data so that more data can stay in memory
.

Kernel Memory Allocation 🔗

Virtual memory is used by the kernel as well as user programs.
The operating system itself must request and release memory, so the kernel has its own memory allocators, such as slab and buddy allocators.
Think of the kernel as a small program inside the operating system. It needs memory too!

Summary 🔗

We have explored the core concepts of virtual memory and how operating systems use memory efficiently through demand paging, page replacement, and copy-on-write.
In the next post, we will look at mass-storage structure and disk scheduling.

References 🔗

    What Is Virtual Memory? — 10 Minutes of Operating Systems, Day 11 | PromleeBlog