principle

The principle of paging storage management ：

Suppose a process size 1KB, We put 1KB Divide into several blocks of the same size , It's called a page Or one page ; Number each page , from 0 Start . At the same time, the memory space is divided into blocks of the same size as the page size , be called block perhaps Page box ; When the process is loaded into memory , In blocks , Divide the process into several parts and load them into non adjacent physical blocks .（ The last page of the process is often not loaded enough to form fragments in the page ）

Basic paging storage management （ Simple paging storage management ） Principle ：

When a job needs to be called into memory , The system first determines whether it can meet the number of blocks required by the process . If it is satisfied, call it into memory , If it is not satisfied, it will not be transferred into memory .

The principle of request paging storage management ：

When the job is called into memory , The system does not need to meet all the needs of the job , You can allocate memory blocks only to the part of the job that you want to run , The rest remains in external storage , When to execute and when to allocate memory blocks .

Page and page table

Page size selection ：

The page size is determined by the address structure of the machine ,32 position /64 position . A machine can only have pages of one size , It's usually a few KB To dozens KB.

There are drawbacks in pages that are too large or too small . Pages are too small to make full use of memory , But page table （ All pages of a job form a page table ） Too long , A lot of memory .

Page too large , Although it can make the page table shorter , Less memory , But there will be a lot of internal debris , Low memory utilization .

Address structure ：

Basic paging storage management （ Simple page storage management ） The implementation of the

Address change mechanism

Two level and multi page tables