[컴퓨터구조] #17. Virtual Memory(1)
Lec 17. Virtual Memory_1 (Large and Fast: Exploiting Memory Hierarchy)
- Virtual Memory
: technique provided by operating systems
• Physical address: address in main memory
• Virtual address: address in virtual space
• Virtual memory: technique that uses main memory as a “cache” for secondary storage
메인메모리를 secondary storage에 대한 cache인 것 처럼 다룸.
- Motivation of Virtual Memory
1) Sharing of memory among multiple programs
복수개의 프로그램을 하나의 메인메모리만 가지고 사용.
2) A single user program may exceed the size of main memory
- Virtual Memory
▪ main memory and secondary storage (HDD)
: Virtual memory automatically manages the 2 levels of memory hierarchy
▪ pages (typically 4KB) : Virtual space is split into fixed-sized blocks
▪ Load only required pages for execution to physical memory
▪ Page fault : a virtual memory miss
▪ Page tables : Operating systems create, contain the translation information
(virtual page -> physical page)
- Two Programs Sharing Physical Memory
OS maps VA to PA by creating page table
- Address Translation
: A virtual address is translated to a physical address by a combination of
hardware(memory management unit) and software
- Page Tables
: OS creates a page table for each process (that is, each program).
VA to PA address translation 정보가 담겨있음. 프로그램마다 존재함.
▪ Page table is located in main memory
▪ Page table is composed of page table entries (PTEs)
▪ PTE is indexed by virtual page number
▪ Page table register in CPU points to page table in main memory
1) requested page is present in main memory
: PTE stores the physical page number plus other status bits
(referenced 참조한 적 있는지,
dirty 내가 갖고있는 메인메모리page내용과 secondary page내용이 같은지 다른지, …)
2) requested page is not present in main memory (page fault)
: PTE can refer to a location in swap space on disk
* swap space : space on the disk reserved for the full virtual memory space of a process (program).
disk라는 커다란 공간에서 내가 실행하려는 프로그램의 각 페이지들을 담고있는 공간.
어떠한 경우에도 특정 virtual page는 swap space에 들어있음이 보장됨.
- Address Translation Mechanism
