What is an Ethernet Switch?

Ethernet switches are switches based on Ethernet to transmit data. Ethernet uses a local area network with a shared bus transmission medium. The structure of the Ethernet switch is that each port is directly connected to the host, and generally works in full duplex mode. The switch can communicate with many pairs of ports at the same time, so that each pair of hosts communicating with each other can transfer data without conflicts like an exclusive communication medium.

Introduced in 1990

(A) VLAN technology

The VLAN technology can effectively avoid malicious attacks on broadcast data. The main reason for malicious attacks on broadcast data is because of the

Ethernet switches are most commonly used, and they are cheaper and complete. Therefore, the application area is very wide, in large and small

1. Each port of the Ethernet switch is directly connected to the host, and generally works at

The Ethernet switch works at the second layer of the OSI network reference model (that is, the data link layer). It is based on MAC (Media Access

As a data transmission device, an Ethernet switch is one of the important devices in a local area network. The internal structure ports are connected to the host. It can connect multiple ports to achieve data transmission without conflict. In addition, the cost of Ethernet switches is low, which can meet the actual needs of different levels. In the context of the big data era, Ethernet switch technology continues to develop and expand to form many complex services. In this process, Ethernet switches also face more serious security issues, mainly including the following aspects: first, malicious broadcast attacks; second, network attacks; third, MAC address attacks; fourth, malicious MAC spoofing Fifth, the loop attack. Take the broadcast malicious attack as an example. The network is an open platform. When receiving a large amount of broadcast data, the switch will forward the data in the form of broadcast. If the transmission control function of the data is not perfect, the network broadband will be affected by these garbage The data is full, and the switch needs to have transmission control functions for many data. ^[3]

1. Cut-through switching

2. Store-and-Forward switching

3. Segment-free switching

Pass-through switching, that is, after the switch receives the frame, as long as it looks at the destination MAC address of the frame, it immediately forwards it to the corresponding port by virtue of the MAC address table; the advantage of this method is that it is fast and the time required for forwarding is short, but The problem is that it is possible to forward some wrong, useless frames to the destination at the same time. The store-and-forward mechanism is that each port of the switch is allocated a certain buffer (memory space, generally 64 k). After the data enters the switch, the target MAC address is read. After understanding the forwarding relationship based on the MAC address table, the data will be It is stored in the buffer of this port until the data fills the buffer and then forwards all the data to the destination at once. While the data is stored in the buffer, the switch will perform a simple check on the data. If incorrect data is found at this time, it will not be forwarded to the destination, but will be discarded here. Of course, this method can provide better data forwarding quality, but the relative forwarding time will be a little longer than the straight-through exchange. Fragment isolation is also called improved straight-through switching. The advantage of using straight-through is that the forwarding delay is small, and the length of each data frame is checked. Because in principle, each Ethernet frame cannot be smaller than 64 bytes and larger than 1518 bytes. If the switch detects frames smaller than 64 bytes or larger than 1518 bytes, it will consider these frames to be "fragmented frames" or "extremely long frames", and then discard them before forwarding. This method combines the advantages of cut-through switching and store-and-forward. Many high-speed switches will use it, but it is not as popular as store-and-forward.

Both pass-through and store-and-forward are a Layer 2 forwarding method, and their forwarding strategies are based on destination MAC (DMAC). There is no difference between these two forwarding methods. The third method is mainly a variant of the first "through-forward".

The biggest difference between them is when they handle forwarding, that is, how the switch handles the relationship between the process of receiving and forwarding the data packet.