1. Write in front

Prior to this, I have learned some knowledge of computer networks. I feel that it is a very basic part. It has not been systematically organized. Therefore, there is no good formation of a relatively complete knowledge system. Recently I have read the content of Jiwang again when I have time, plus learned networking technology, "learning the new through reviewing the old", and discovering some knowledge and content that have not been learned before, especially the knowledge in the latter part, and I feel that I have gained a lot of rewards. . I am going to record this time. On the one hand, I can better summarize the important knowledge points, and on the other hand, it is convenient for viewing and learning.

2. Internet and Internet

As the name implies, a network is composed of several nodes and links connecting these nodes. The concept of nodes is extensive, and devices such as computers, routers, and switches can all become nodes in the network.

Internet (Internet) is a general term that refers to a network formed by interconnecting multiple computer networks.
Internet (Internet) is a term that refers to the world's largest and open specific computer network consisting of many interconnected networks. It uses the TCP/IP protocol suite and was formerly known as ARPANET in the United States.

The development of the Internet is divided into three major stages:

The first stage: From a single network ARPANET to the Internet, the embryonic form of the Internet was born.

The second stage: the establishment of a three-level structure of the Internet, divided into backbone network, regional network and campus network (or enterprise network).

The third stage: the formation of a multi-level ISP (Internet Service Provider, Internet Service Provider) structure of the Internet.

It can be seen that the Internet is developing rapidly, so there needs to be a standard to unify at this time, so as to avoid the incompatibility of technical systems. Therefore, international standards were born and are fully managed by the international organization ISOC.

After this, all Internet standards are published on the Internet in the form of RFC documents. RFC (Request For Comments) : The meaning of requests and comments. The document numbers are named in chronological order: RFC xxxx, for example, [RFC 792] is the number of the Internet standard of ICMP.

The formulation of official standards for the Internet has to go through 4 stages:

Internet Draft : This stage is not yet an RFC document.

Proposed standard : From this stage, it becomes an RFC document.

Draft standard .

Internet standards .

3. The core of the Internet: circuit switching-message switching-packet switching

The core part of the Internet is the most complex part. It needs to provide connectivity to ensure that any host on the edge of the network can communicate with each other.

The key to this core is the router . The router is a key component in the continuous network development and realizes packet switching. Prior to this, it has gone through the historical stages of circuit switching and message switching.

1. Circuit switching

In the era when circuit switching began with the advent of telephones, the communication between telephones needed to function through an intermediate telephone switch. When two phones communicate, a dedicated physical path is established : establish a connection (allocate communication resources) ---> call (always occupy resources) ---> release the connection (return resources) .

It can be seen that this method only allows two phone users to communicate at a time, because the entire channel is monopolized, and resources are released only when the call ends. When applied to computer communication to transmit data, the efficiency is obviously very low, and the communication line is idle most of the time.

Circuit switching can be seen as the bit stream of data transmission reaching the end point.

2. Message exchange

The message exchange transmits all the data to the next node of the link at one time, stores them all and searches for the next node, and finally forwards them to the destination at one time.

3. Packet switching

In packet switching, the message is not transmitted as a whole, but divided into multiple smaller data segments, and the packet is forwarded through the router. The reason for the high efficiency is that the data segment can be stored in the memory instead of being written to disk. Speed up the exchange rate.

Packet switching is to transfer a single packet data to the next node, and the output interface and input interface can work in parallel.

The advantages of packet switching are high efficiency (dynamically allocate bandwidth, occupying links section by section), flexibility (select routers for each packet), rapid (using small packets as the transmission unit), and reliability (distributed multi-route packet switching network, survival performance Great).

4. The compromise between OSI and TCP/IP: five-layer protocol principle architecture

We all know that the OSI seven-layer protocol is the real international standard. But before the OSI model was put forward, the TCP/IP protocol suite had been widely used and was the de facto standard.

The main reasons are: (1) The OSI model standard has no commercial driving force. (2) The implementation of the OSI protocol is too complicated and the operating efficiency is low. (3) The OSI standard equipment did not enter the market in time, and the formulation cycle was too long. (4) The division of OSI levels is unreasonable, and some functions appear repeatedly.

The OSI system structure is: (from bottom to top) the physical layer, data link layer, network layer, transport layer, session layer, presentation layer, and application layer.

As can be seen from the above figure, the TCP/IP system structure is divided into 4 layers. In fact, only the upper three layers function, and the bottom network interface layer has no content.

Therefore, integrating the advantages of the lower three layers of OSI and the upper three layers of TCP/IP, a five-layer protocol architecture is adopted.

The functions of each layer are:

Application layer : the highest layer of the architecture. The application of a specific network is completed through the interaction between application processes. This layer of protocols includes e-mail SMTP protocol, file transfer FTP protocol, etc., and the data unit is a message .

Transport layer : The service responsible for data transmission between two host processes. The transport layer has functions of demultiplexing and multiplexing. Reuse means that multiple application layer processes from top to bottom can use the services of the transport layer at the same time, and demultiplexing means that the bottom-up transport layer delivers the received data to the corresponding process in the application layer. The two major protocols of this layer are TCP, the data unit is message segment; UDP, the data unit is user datagram . The specific content of TCP and UDP is planned to be recorded in detail in a later article.

Network layer : Responsible for providing communication services for different hosts in the packet switching network. The data unit is an IP datagram . The router is the protagonist of this layer, the task is to route and forward .

Data link layer : Responsible for transmitting packets from one end of the link to the other end, and the data unit is a frame .

Physical layer : The bottom layer of the system structure, which completes the most basic tasks in a computer network. The data unit is a bit stream . The physical transmission intermediary includes twisted pair, coaxial cable, optical cable, etc.

Five, write at the back

In this article, I mainly wrote some overviews of computer networks. The more important one is the five-layer protocol architecture, and the TCP/IP protocol suite is a key part of it. Therefore, in the following content records, I should study and summarize the specific knowledge of the five-layer protocol. The main plan is to start from the data link layer and go up. I personally feel that the physical layer is biased towards knowledge of communication principles, including channels, coding, and modulation. Knowledge is not unfolded.

If you think it's good, welcome to "one-click, three-link", like, bookmark, follow, comment directly if you have any questions, exchange and learn!

My CSDN blog: https://blog.csdn.net/Charzous/article/details/115270122

Computer Network Study: TCP/IP Principle and Architecture