Chapter 10 – Paging, Page Tables & Swap

Core idea

The OS uses paging to manage memory efficiently and transparently.

Page tables

• Map virtual page → physical frame
• Stored in memory
• CPU caches translations in TLB

Swap

• When RAM is full:
  • Unused pages moved to disk (swap)
• On access:
  • Page fault
  • OS loads page back into RAM

Costs

• Page faults are expensive
• Disk ≪ RAM speed

Exam traps

• Swap ≠ more RAM (it’s slower)
• TLB misses hurt performance