// The computer connected to the network is called the host, which can be a computer or other intelligent terminals
Official Internet Standard
The draft Internet (Internet Draft) - valid for only six months . At this stage also is not RFC documents.
Proposed Standard (Proposed Standard) - From the beginning of this stage becomes the RFC documents.
Internet standards (Internet Standard) - After achieving formal standards, each standard on a number assigned to STD xxxx . A standard can be associated with multiple RFC documents.
In addition to the two RFC documents of the proposed standard and the Internet standard , there are three kinds of RFC documents, namely historical, experimental and informative RFC documents .
The composition of the Internet
From the perspective of the way the Internet works, it can be divided into two major blocks:
(1) Edge part: It consists of all hosts connected to the Internet. This part is used directly by users to communicate (transfer data, audio or video) and share resources.
(2) ** The core part: ** consists of a large number of networks and routers connecting these networks. This part provides services for the edge part (providing connectivity and switching).
Logically, the part that constitutes the equipment and links that provide connectivity and exchange of public services is the core part.
The part on the edge of the Internet is all the hosts connected to the Internet. These hosts are also called end systems.
Substantial communication between computers: "Host A and Host B communicate" actually refers to: "running on the host A a program running on the host and B another program on the communication" .
Two communication methods between end systems
The communication methods between end systems can generally be divided into two categories:
Client - server method ( C / S method)
I.e. Client / Server mode, referred to as C / S mode.
Peer-to-peer ( P2P )
That is Peer - to - Peer mode, referred to as P2P mode.
CS architecture
The client software ** is run after being called by the user, ** initiates communication (requesting service) to the remote server when it intends to communicate. Therefore, the client program must know the address of the server program. No special hardware and very complicated operating system are required.
Server software is a program used to provide a certain service, and can simultaneously process multiple remote or local client requests. After the system is started, it is automatically called and continuously running. It passively waits and accepts communication requests from customers everywhere. Therefore, the server program does not need to know the address of the client program . Generally requires powerful hardware and advanced operating system support.
After the communication relationship between the client and the server is established, the communication can be bidirectional, and both the client and the server can send and receive data
P2P
Peer connection (Use the peer-to-Use the peer , abbreviated as P2P) means that the host does not distinguish between the two which is a communication service provider or a service requester.
As long as both hosts run peer-to-peer connection software (P2P software ) , they can communicate on an equal, peer-to-peer connection.
Both parties can download shared documents that the other party has stored in the hard disk.
The peer-to-peer connection is still essentially using the client-server method, but each host in the peer-to-peer connection is both a client and a server
The peer-to-peer connection work mode can support a large number of peer users (such as millions) to work at the same time.
The more hosts participating in p2p communication, the faster the system may be?
The more users participating in the P2P network, the more uploaders there are, and the more uploaders there are, the less the opportunity for P2P terminals to fetch data from the server, so the existence of P2P not only greatly reduces the pressure on the server The more the number of people, the faster the effect.
Three ways of information exchange
The core part of the network is the most complicated part of the Internet.
The core part of the network should provide connectivity to a large number of hosts at the edge of the network, so that any host in the edge can communicate with other hosts (that is, transmit or receive various forms of data).
The router in the core part of the network plays a special role . The router is a key component to implement packet switching . Its task is to forward the received packets. This is the most important function of the core part of the network.
The switching itself is to save resources. If switching is not used, n (n-1) / 2 lines are required between n terminals.
What is exchange
Here, the " exchange" (Switching) meaning that the adapter - the phone line transferred to another telephone line, so that they communicate together.
From the perspective of communication resource allocation, "switching" is to dynamically allocate transmission line resources in a certain way .
Circuit switched features
It must be a circuit-switched connection-oriented to.
Circuit switching is divided into three stages:
Establishing a connection: establishing a dedicated physical channel , to ensure that communication resources required at both the call can not be occupied by another user communication;
Communication: Both the calling and called parties can talk to each other ;
Release connection: release the dedicated physical channel just used ( release all the communication resources just occupied).
Disadvantages: Computer data is bursty. This leads to a very low utilization rate of communication lines when transferring computer data ( the time used to transfer data is often less than 10% or even 1% ).
Main features of packet switching
Packet switching uses store-and-forward technology .
At the sending end, first divide the longer message into shorter, fixed-length data segments. It is assumed that this long insulation is not convenient for transmission.
-
Add a header component packet: add a header component packet (packet) before each data segment.
-
Adding a segment on the front of each header portion constituting the packet (packet). Packet switching networks use "packets" as data transmission units.
Send each packet to the receiver in turn (assuming the receiver is on the left). -
The importance of the packet header: The header of each packet contains control information such as addresses (such as destination and source addresses).
The packet-switched network forwards the packet to the next node switch based on the address information in the received packet header.
Each packet independently selects the transmission path in the Internet.
With this store-and-forward approach, the final packet can reach its final destination. -
Strip the header after receiving the packet: the receiving end strips the header after receiving the packet and restores it as a message.
-
Finally, it is restored to the original message: Finally, the received data is restored to the original message at the receiving end.
Here we assume that the packet has no errors during transmission and has not been discarded during forwarding.
Problems caused by packet switching:
Packets need to be queued when stored and forwarded at each node, which will cause a certain delay.
The header that the packet must carry (which contains essential control information) also causes a certain amount of overhead.
router
There is no direct connection between the input and output ports in the router.
The process of processing packets by the router is:
- Put the received packet into the buffer first (temporary storage);
- Look up the forwarding table and find out from which port a certain destination address should be forwarded;
- Forward the packet to the appropriate port.
The host performs information processing for users, and sends packets to the network, and receives packets from the network.
The router stores and forwards the packet, and finally delivers the packet to the destination host.
The delay in the exchange process comes from the processing time.
Summary of three exchange methods
If a large amount of data is to be continuously transmitted and the transmission time is much longer than the connection establishment time, the transmission rate of circuit switching is faster.
Since the length of a packet is often much shorter than the length of the entire message, packet switching has a smaller delay than message switching, and it also has better flexibility .
