Principles of computer composition (3) - storage system

Series Article Directory

        Chapter 1 Computer System Overview

        Chapter 2 Representation and Operation of Data

        Chapter 3 Storage System

        Chapter 4 Command System

        Chapter 5 CPU

        Chapter 6 Bus

        Chapter 7 Input/Output System


Chapter 3 Storage System

Series Article Directory

1. Memory overview

1.1 Classification of memory

1.2 Memory Performance Indicators

1.3 Hierarchical structure of memory

2. Simple model of main memory and addressing method

3. Semiconductor Random Access Memory

3.1 SRAM and DRAM

3.2 Read Only Memory ROM

Fourth, the connection between the main memory and the CPU

4.1 Main memory capacity expansion - bit expansion

4.2 Main memory capacity expansion - word expansion 

4.3 Examples

5. Dual-port RAM and multi-module memory

5.1 Dual-port RAM

5.2 Multi-block memory

6. Cache memory

6.1 The principle of locality of program access

6.2 The basic working principle of Cache

6.2.1 Cache-main memory address mapping

6.2.2 Replacement Algorithm for Main Memory Blocks in Cache

6.2.3 Cache write back strategy

6.3 Examples

Seven, virtual memory 


1. Memory overview

1.1 Classification of memory

In addition to being divided according to the function level, the memory can also be classified according to the access method:

  • Random access memory (RAM), the content of any storage unit of the memory can be accessed randomly, and the access time has nothing to do with the physical location of the access unit. It is mainly used as main memory or cache memory, and the content will be lost after power off.

  • Read-only memory (ROM), the contents of the memory can only be read randomly and cannot be written, even if the power is turned off, the contents will not be lost.

  • Serial access memory, when performing read/write operations on storage units, it needs to be addressed according to the order of its physical location, including sequential access memory (such as tape) and direct access memory (such as disk).

1.2 Memory Performance Indicators

1.3 Hierarchical structure of memory

Summary (to sort out the knowledge points in this section!!!) :


2. Simple model of main memory and addressing method

The role of each component in the memory chip structure:

  • The memory matrix is ​​composed of a large number of identical bit memory cell arrays.
  • The decoding drive translates the address signal from the address bus into the gate signal corresponding to the storage unit, and the signal completes the read/write operation to the selected unit with the cooperation of the read-write circuit.
  • Read and write circuits, including sense amplifiers and write circuits, are used to complete read/write operations.
  • The read/write control line determines the read/write operation of the chip.
  • The chip selection line determines which memory chip is selected and can be used for capacity expansion.
  • The address line is a one-way input, and its number of bits is related to the number of stored words.
  • The data line is bidirectional, and its number of bits is related to the number of bits of data read or written.


3. Semiconductor Random Access Memory

3.1 SRAM and DRAM

3.2 Read Only Memory ROM

Summary (to sort out the knowledge points in this section!!!):


Fourth, the connection between the main memory and the CPU

4.1 Main memory capacity expansion - bit expansion

4.2 Main memory capacity expansion - word expansion 

Summary (to sort out the knowledge points in this section!!!):

4.3 Examples


5. Dual-port RAM and multi-module memory

According to Moore's Law, the performance of integrated circuits doubles every 18 bits. For the CPU, the speed increases exponentially , and for the main memory, its storage capacity increases exponentially, but its storage speed is close to linear growth , which will lead to CPU and The speed gap of the main memory is getting bigger and bigger, so how to improve the working speed of the main memory has become a major topic of optimization in the computer field.

The dual-port RAM and multi-module memory that will be introduced in this section are to achieve the purpose of improving the working speed of the memory by using the management structure of the special memory.

5.1 Dual-port RAM

5.2 Multi-block memory

Interleaved memory is actually a modular memory that can perform multiple independent read/write operations in parallel. 

Summary (to sort out the knowledge points in this section!!!):


6. Cache memory

6.1 The principle of locality of program access

6.2 The basic working principle of Cache

6.2.1 Cache-main memory address mapping

6.2.2 Replacement Algorithm for Main Memory Blocks in Cache

6.2.3 Cache write back strategy

6.3 Examples


Seven, virtual memory 

The content about virtual storage is introduced in detail in Chapter 3 of the operating system , link: Summary of Operating System Knowledge Points 

The following is just a brief introduction

5. Dual-port RAM and multi-module memory ------- Improve the working speed of the memory
6. Cache cache memory ------- Improve the working speed of the storage system
7. Virtual memory------ - Increase the capacity of the storage system 

Guess you like

Origin blog.csdn.net/weixin_52850476/article/details/125619516