Let's talk about page file transfer from the perspective of data package
A packet of “ Journey ” The following are from “ How the packet reaches the host ”“ How hosts deliver packets to applications ” and “ How data is fully delivered to the application ” These three perspectives will tell you the process of data transmission .
Internet , In fact, it's a set of architecture composed of concepts and protocols . among , The protocol is a well-known set of rules and standards , If all parties agree to use , Then the communication between them will become unimpeded .
Data in the Internet is transmitted through packets . If the data sent is large , Then the data will be split into many small packets for transmission . Like the audio data we're listening to now , It is split into small packets for transmission , It's not a big file that's transferred at one time .
1.IP: Send the packet to the destination host
Packets are transmitted over the Internet , It's about conforming to the internet protocol (Internet Protocol, abbreviation IP) standard . Different online devices on the Internet have unique addresses , The address is just a number , This is similar to most home delivery addresses , Just need to know the specific address of the family , You can send the package to this address , So the logistics system can deliver the goods to the destination .
The address of the computer is called IP Address , Visiting any website is really just your computer asking for information from another computer .
If you want to send a packet from the host A Send to host B, So before transmission , The packet will be attached to the host B Of IP Address information , In this way, it can be properly addressed during transmission . Extra , A host will be attached to the packet A Of itself IP Address , With these information hosts B Can reply to the host A. These additional information will be put into a file called IP In the data structure of the header .IP The head is IP Information at the beginning of the packet , contain IP edition 、 Source IP Address 、 The goal is IP Address 、 Time to live and so on .
2.UDP: Deliver the packet to the application
IP It's a very low-level protocol , It is only responsible for transferring data packets to the other party's computer , But the other computer doesn't know which program to give the data package to , To browser or to King glory ? therefore , Need to be based on IP Develop protocols that work with applications , The most common is “ User packet protocol (User Datagram Protocol)”, abbreviation UDP.
UDP One of the most important messages in this project is Port number , The port number is actually a number , Every program that wants to access the network needs to bind a port number . Through port number UDP Can send the specified packet to the specified program , therefore IP adopt IP The address information sends the packet to the designated computer , and UDP Distribute the data package to the correct program through the port number . and IP The head is the same , The port number will be loaded into UDP Inside the head ,UDP The header is combined with the original data package to form a new UDP Data packets .UDP Except for the destination port in the header , And the source port number .
In the use of UDP When sending data , There are various factors that can cause packet errors , although UDP It can check whether the data is correct , But for the wrong packets ,UDP There is no retransmission mechanism , Just discard the current package , and UDP It's impossible to know whether the destination can be reached after sending .
although UDP There is no guarantee of data reliability , But the transmission speed is very fast , therefore UDP Will be applied to some attention speed 、 But areas where data integrity is less stringent , Like online video 、 Interactive games, etc .
3.TCP: Deliver the data completely to the application
For browser requests , Or mail, which requires data transmission reliability (reliability) Application , If you use UDP There are two problems with coming to transport :
- Data packets are easily lost in the process of transmission ;
- Large files will be split into many small packets for transmission , These small packets will go through different routes , And at different times to the receiving end , and UDP The protocol doesn't know how to assemble these packets , To restore these packets to a complete file .
Based on these two questions , We introduce TCP 了 .TCP(Transmission Control Protocol, Transmission control protocol ) It's a connection oriented 、 reliable 、 Transport layer communication protocol based on byte stream . be relative to UDP,TCP There are two characteristics :
- In the case of packet loss ,TCP Provide retransmission mechanism ;
- TCP The packet sorting mechanism is introduced , It is used to ensure that the disordered packets are combined into a complete file .
and UDP The head is the same ,TCP The header contains the target port and the native port number , The serial number for sorting is also provided , So that the receiving end can rearrange the data packet through the serial number .
TCP The transmission process of a single packet and UDP The process is similar , The difference is this , adopt TCP Header information ensures the integrity of a large piece of data transmission .
Let's take a look at the whole TCP Connection process , Through this process you can understand TCP How to ensure the retransmission mechanism and packet sorting function .
As can be seen from the figure below , A complete TCP The life cycle of the connection includes “ Establishing a connection ”“ To transmit data ” and “ disconnect ” Three stages .
- First , Establish connection phase . This stage is through “ Three handshakes ” To establish a connection between the client and the server .TCP Provide connection oriented communication transmission . Connection oriented refers to the preparation between two ends before data communication . So-called Three handshakes , It means building a TCP When the connection , The client and server send three packets to confirm the connection ( for the first time : The client sends a message with SYN Packets for flags The second time : The server sends a message with SYN and ACK Marked packets third time : The client sends a message with ACK Packets for flags ).
- secondly , Data transmission phase . In this phase , The receiver needs to confirm every packet , That is to say, after receiving the packet, the receiver , Need to send confirmation packet to sender . So when the sender sends a packet , The confirmation message fed back by the receiver is not received within the specified time , Then it is judged that the packet is lost , And trigger the retransmission mechanism of the sender . Again , A large file will be split into many small packets during transmission , When these packets arrive at the receiver , The receiver will follow TCP The sequence number in the header is the order , So as to ensure the composition of complete data .
- Last , Disconnection phase . After data transmission , It's about to terminate the connection , It's about the last stage “ Four waves ” To ensure that both sides can disconnect ( for the first time : The client sends a message with FIN Packets for flags The second time : The server receives FIN After package , Send to client with ACK Data packets third time : The server sends a FIN To the client The fourth time : Client received FIN after , Send to the server with ACK Packets for flags ).
Here you should understand ,TCP In order to ensure the reliability of data transmission , At the expense of packet transmission speed , because “ Three handshakes ” and “ Packet verification mechanism ” And so on to double the number of packets in the transmission process .
summary
- Data in the Internet is transmitted through packets , Packets are easy to lose or make mistakes in the process of transmission .
- IP Responsible for delivering packets to the destination host .
- UDP Responsible for delivering packets to specific applications .
- and TCP Ensure the complete transmission of data , Its connection can be divided into three stages : Establishing a connection 、 Transfer data and disconnect .
Reference material