Network Protocol Chapter 1 _ Overview

Chapter 1 _ overview of network protocols are generally divided into different layers for development. TCP/IP is generally considered to have four layers, from top to bottom: Application Layer: processing applications, common TCP/IP applications include Telnet (Remote logon), FTP (file transfer protocol), SMTP (Simple Mail Transfer Protocol), and SNMP (Simple Network Management Protocol) transport Layer: Provides end-to-end communication for applications on two hosts. The Transport Layer mainly includes two Protocols: TCP (Transport Control Protocol) and UDP (User Datagram Protocol) Network Layer: process Group Activities in the network. Network Layer Protocols include IP protocol (Internet Protocol), ICMP protocol (Internet Control Packet protocol), IGMP protocol (Internet Group Management Protocol) Link Layer: network interface layer, it usually includes the device drivers in the OS and the corresponding network interface card in the computer. The application layer of the Physical Interface Details of the instrument processing and cable is generally a user process, and the lower layer is generally executed in the OS kernel. The application layer processes users' applications, the link layer processes communication-related media details, and the application and transmission layers use end-to-end protocols, the network layer provides an end-to-end and point-to-point explanation ( http://www.bkjia.com/net/201305/209580.html Point-to-point is a physical topology. For example, a fiber optic cable must be a point-to-point connection, and a ddnleased line is also used. That is, no machine exists in the middle of each machine on both ends. Point-to-point is at the network layer. Your transport layer only thinks that my data is directly sent from a to e, but this is not the case. For example, the transport layer is like a leader, and he releases the command: what do you want to do, but what you really do is not him. What you really do is an employee. Maybe the leaders think that you can do a good job in a very simple sentence, but it is difficult for employees to enter the sky, and the procedures are extremely cumbersome, therefore, the transport layer is the leader of the command release, and he is talking about the command, that is, the ultimate goal. Therefore, he only sees the initial address and the final address, both the two endpoints of a task, the network layer is equivalent to an employee. I need to complete the tasks of leadership step by step, from a to B, from B to c ..., so what he sees is only a phase of the entire task, a to B, B to c... this is point-to-point. End-to-end is a network connection. To communicate with the network, you must establish a connection. No matter how far there are, and how many machines are in the middle, you must establish a connection between the two ends (source and destination). Once the connection is established, it is already an end-to-end connection, that is, the end-to-end is a logical link. This path may go through a complex physical route. However, no matter whether the two hosts are at either end, they only think that there is a connection between the two ends, once the communication is complete, the connection is released, and the physical line may be used by other applications to establish a connection. TCP is a specific protocol used to establish such an end-to-end connection. So is SPX. The end-to-end is the transmission layer. For example, if you want to transmit data from A to E, it may go through A-> B-> C-> D-> E, for the transport layer, he does not know the existence of B, c, and d. He only thinks that my packet data is directly from a to e, which is called end-to-end. In a word, end-to-end is implemented and composed of numerous point-to-point implementations. The services provided by the network layer IP address are not reliable, but TCP provides a reliable transmission layer by transmitting and receiving end-to-end validation grouping mechanisms through timeout retransmission. Bridges Interconnect Networks at the link layer, and routers Interconnect Networks at the network layer. TCP/IP tends to use the 32bitIP address of the router Internet: Class A: 0 + 7bit network number + 24bit host number, range: 0.0.0.0-127.255.255.255 Class B: 10 + 14bit network number + 16bit host number, range: 128.0.0.0-191.0000255 Class C: 110 + 21bit network number + 8bit host number, range: 192.0.0.0-223.0000255 D class: 1110 + 28bit multicast group number, range: 224.0.0.0-239.00000000255 E Class: 11110 + 27bit (retained), range: 240.0.0.0-247.0000255 sub-assembly: TCP and UDP use a 16-bit port number to indicate different applications. TCP and UDP respectively store the source port number and destination port number into the packet header. Each layer of the Protocol box should check the protocol ID in the packet header to determine the upper layer protocol for receiving data. This process is called sub-use. Customer-Server Model: Server-specific classification: duplicate type and concurrent hairstyle. Generally, the TCP server is concurrent and the UDP server is duplicate. The server uses a well-known port number to identify the program (telnet uses port 23, ftp is 21, and tftp is 69 ). Well-known port numbers range from 1 ~ Between 255 and 256 ~ 1023 is occupied by unix systems. DNS provides a dynamic ing between the host name and IP address. The port number is used to identify applications that communicate with each other. The server uses a well-known port number, and the client uses a temporary port number. The maximum difference between the network layer and the transport layer is: the network layer provides point-to-point services, and the transport layer provides end-to-end services.