View the development history of network switch technology

The network switch technology allows you to adjust the shared and dedicated LAN segments, which effectively reduces the information flow bottleneck in the LAN and ensures the security and stability of our network. In 1993, LAN switching equipment emerged. In 1994, the domestic network switch technology boom was set up. In fact, the network switch technology is a simplified, low-cost, high-performance and high-port intensive switch product, reflecting the complex switch technology of the bridge technology in the OSI reference model layer 2 operation.

Like the bridge adapter, the switch simply decides to forward information based on the MAC address in each package. This type of forwarding decision generally does not consider the deeper information hidden in the package. The difference with the bridge adapter is that the forwarding delay of the switch is very small, and the operation is similar to the performance of a single LAN, far exceeding the forwarding performance between common Bridging networks. The network switch technology allows you to adjust the bandwidth of shared and dedicated LAN segments to reduce the bottleneck of information flow between LAN. Now there are Ethernet, fast Ethernet, FDDI and ATM exchange products.

Similar to traditional bridges, vswitches provide many network interconnection functions. Vswitches can economically divide networks into small conflicting domains to provide higher bandwidth for each workstation. Protocol transparency allows the switch to be directly installed in a multi-protocol network with simple software configuration. The switch uses the NIC of the existing cable, repeater, hub, and workstation, you do not need to perform high-level hardware upgrades. vswitches are transparent to workstations. This reduces management costs and simplifies the operations of adding, moving, and changing network nodes.

Using a specially designed integrated circuit, the switch can forward information on all ports at a line rate in parallel, providing much higher operation performance than the traditional bridge. For example, theoretically, a single Ethernet port can provide a transmission rate of 14880bps for data packets containing 64 octal packets. This means that a "line rate" Ethernet switch with 12 ports that support six parallel data streams must provide a total throughput of BPS and six information flows (X14880bps/information flows ). The dedicated integrated circuit technology enables the switch to run with more ports. The port cost is lower than that of the traditional bridge.

Several network switch technologies

1. Port Switching

Port switching technology first appeared in a slot hub. The backplane of a hub is usually divided into multiple Ethernet segments, each of which is a broadcast domain, networks do not communicate with each other. After the master module is inserted, it is usually allocated to the network segment of a backplane. Port switching is used to allocate and balance the Ethernet module ports among multiple network segments of the backplane. Based on the degree of support, port switching can also be subdivided:

◆ Module switching: migrate the entire module to the network segment.

◆ Port group switch: ports on the module are usually divided into several groups, and each group of ports allows network segment migration.
 
◆ Port-level switch: each port can be migrated between different network segments. This switching technology is based on the first layer of OSI and has the advantages of flexibility and load balancing capabilities. If properly configured, fault tolerance can still be implemented to a certain extent, but the characteristics of the shared transmission media are not changed, and thus cannot be called a real exchange.

2. Frame Switching

Frame switching is currently the most widely used lan network switch technology. It provides a parallel transmission mechanism to reduce conflicting domains and obtain high bandwidth by performing differential segments on Traditional transmission media. Generally, the network switch technology of each company's products is different, but the network frame processing methods are generally as follows:

◆ Pass-through switching: Provides line rate processing capabilities. The switch only reads the first 14 bytes of the network frame and then transfers the network frame to the corresponding port.

◆ Storage and forwarding: checks and controls the reading of network frames.
 
The exchange speed of the previous method is very fast, but it lacks more advanced control over network frames, lacks Intelligence and Security, and cannot support port exchange with different rates. Therefore, each vendor focuses on the latter technology. Some manufacturers even break down network frames and break down frames into fixed cells. This cell processing is easy to implement by hardware, and the processing speed is fast, at the same time, advanced control functions such as LET hub of MADGE can be completed, such as priority control.

3. Cell Exchange

ATM technology uses a fixed length of 53 bytes of Cell Exchange. Because the length is fixed, it is easy to implement with hardware. ATM uses dedicated non-differential connections and runs in parallel. multiple nodes can be established through a switch at the same time, but the communication capability between each node is not affected. ATM also allows multiple virtual links at the source node, target node, and node to ensure sufficient bandwidth and fault tolerance. The statistical time-division circuit is used in ATM, which can greatly improve the channel utilization. The bandwidth of an ATM can reach 25 M, 155 M, M, or even several Gb. However, with the emergence of 10-ge Ethernet, the ATM technology, which once represents the future direction of the development of the network and network switch technology, began to lose its meaning.