TCP/IP detailed--Data link layer __ Software--Network--basic knowledge of network

The most common encapsulation on the link layer is Ethernet and IEEE 802 encapsulation, where Ethernet is used as an object of instruction.

I. Overview the data link layer belongs to the bottom of the computer network, and the channel used mainly has two kinds of point-to-point channel and broadcast channel.
In the TCP/IP protocol family, the data link layer has three main purposes: 1 send and receive data for the IP Module 2, send ARP request and receive ARP reply 3 for the ARP module, send RARP request and receive Rarp reply for the Rarp module            TCP/IP supports many different link layer protocols, depending on the hardware used by the network, such as Ethernet, Token Ring, FDDI (Fiber Distributed Data Interface) and RS-232 serial lines.      Data Link Layer Protocol data Unit-frame: the IP layer (network layer) of the datagram add header and tail encapsulated into frames. There are many kinds of data link layer protocols, which can solve three basic problems, encapsulation into frames, transparent transmission, error detection.

Ii. Ethernet and IEEE802 encapsulation as shown in the illustration, shows two different forms of encapsulation format. The most commonly used encapsulation format is the RFC894-defined format. Both frame formats use the destination address and the source address of 48bit (6 byte).

The Ethernet frame format has only 6 bytes of destination address and source address. The frame length field refers to the byte length of its subsequent data, but does not include the CRC checksum code. This length is the data payload that is provided to the upper layer protocol, that is, the maximum length of the data in the upper protocol package does not exceed the data length of the Ethernet frame.         Minimum data frame tail 46, maximum data frame length of 1500. In the analysis of the length of the Ethernet frame, there are different statements, see http://blog.csdn.net/yusiguyuan/article/details/22490229.
Second, Maximum transmission unit MTU
The Maximum transmission unit (Maximum transmission UNIT,MTU) refers to the maximum packet size (in bytes) that can be passed on a layer of a communication protocol. If the IP layer to transmit a datagram than the link layer of the MTU is larger, then the IP layer will be a slice of the datagram. A datagram is divided into slices, each of which is less than or equal to the MTU value of the link layer. When hosts on the same network communicate with each other, the MTU of the network is important to both sides of the communication. However, when the host to communicate through a lot of networks, the most important to both sides of the communication path is the smallest MTU, because the communication path on the different network link layer MTU different.      The smallest MTU in a communication path is called the Path MTU. The default values for some common link layer protocol MTU in the network are as follows: FDDI protocol: 4352-byte Ethernet (Ethernet) protocol: 1500-byte PPPoE (ADSL) protocol: 1492-byte X.25 protocol (Dial up/modem): 576 bytes point-to-point:4470 byte Ethernet and 802.3 have a limit on the length of the data frame, with a maximum value of 1500 and 1492 bytes respectively. This feature of the link layer is called the MTU, the maximum transmission unit, and most of the different types of networks have an upper bound, which is determined by the hardware.

If the IP layer has a datagram to pass, and the length of the data than the link layer of the MTU is larger, then the IP layer needs to be fragmented, the data into several slices, so each piece is less than the MTU.
Third, Path MTU
The path MTU is similar to the above content, except that you need to know the minimum MTU of the entire path when transferring data.