Summary of the agreement

1.1 Layering

link Layer (Data Link layer/network interface layer): includes device drivers in the operating system and computers for network interface cards, together with the physical interface information of the cable (or other transmission medium)

Network layer (Internet layer): handles packet activity, including IP protocol, ICMP protocol, IGMP protocol

Transport Layer: provides end-to-end communication for applications on the host, including TCP and UDP protocols;

TCP provides reliable data communication, that is, the application layer of data into small pieces to the following network layer, confirm the received packet, set to send the last confirmation packet of the super clock;

UDP provides a simple service, sending datagrams to the host only, does not guarantee the reliability

Application layer: handles specific application details such as telent (Telnet), FTP (File Transfer Protocol), SMTP (Simple Mail Transfer Protocol), SNMP (Simple Network Management Protocol)

1.2 Layered protocol for TCP/IP

1.3 Internet Address

1.4 Package

1.5 cent use

1.6 Customer-Server model

Duplicate servers:

I1. Wait for the arrival of a customer request.
I2. Process customer requests.
I3. Sends a response to the customer who sent the request.
I4. Returns I 1 steps.
　　
and hairstyle server:
C1. Wait for the arrival of a customer request.
C2. Start a new server to handle requests from this client. During this time a new process, task, or thread may be generated and dependent on the support of the underlying operating system. How this step is done depends on the operating system. The new server that is generated processes all requests from the customer. After processing is finished, terminate this new server.
C3. Return to step c 1.

The primary problem with duplicate servers occurs in the I 2 state. At this point, it cannot serve other clients. The advantage of a concurrent server is that it uses the method of generating additional servers to process customer requests. That is, each customer has its own corresponding server. If the operating system allows multitasking, it can serve multiple customers at the same time.

1.7 Port number

(1) Recognized port (well known Ports): This type of port is also often referred to as a "common port". The port numbers of such ports are from 0 to 1024, and they are tightly bound to certain services. Typically, the communication of these ports clearly indicates the protocol for a service, which is no longer a way to redefine its role. For example: Port 80 is actually always used for HTTP communication, while Port 23rd is dedicated to the Telnet service.

(2) Register port (registered Ports): Port number from 1025 to 49151. They are loosely tied to some services. It is also said that there are many services bound to these ports, which are also used for many other purposes. Most of these ports do not explicitly define service objects, and different programs can be defined on their own.

(3) dynamic and/or private ports (and/or private Ports): Port numbers from 49152 to 65535. In theory, common services should not be allocated on these ports. In fact, some of the more special programs, especially some Trojan horse programs like to use these ports, because these ports are often not attracted attention, easy to hide.

The port can also be classified as "TCP protocol Port" and "UDP protocol Port", depending on the mode of service provided.

Note that this number is used on the tcp,udp of a logical number, not a hardware port, we usually say that a certain port sealed off, but also only in the IP layer with this number of IP packets to filter out just.

1.8 Application Programming Interface

Applications using the TCP/IP protocol use two programming interfaces (APIs): Socket and TLI (Transport Layer Interface)

Summary of the agreement