In 1972, Bob Metcalfe, "the father of Ethernet", was hired by Xerox as a network expert. Bob Metcalfe's first task when he came to Xerox's Palo Alto Research Center (PARC) was to connect Palo Alto's computers to ARPANET (the predecessor of the Internet). superior. At the end of 1972, Bob Metcalfe designed a network based on the ALOHA system (a radio network system) and named it the ALTO ALOHA network. The network was renamed Ethernet in 1973. This was the original Ethernet experimental prototype. The network transmission medium is thick coaxial cable with a rate of 2.94Mbps. At the end of 1977, Bob Metcalfe and three of his collaborators obtained the patent for "Multipoint Data Communication System with Collision Detection". The multipoint transmission system was called CSMA/CD (Carrier Sense Multiple Access with Collision Detection). Multiple access/conflict detection) system. From then on, Ethernet was officially born.
In 1979, DEC (Digital Equipment Corporation), Intel Corporation and Xerox Corporation (DIX Alliance) formed an alliance to promote the standardization of Ethernet. In 1982, the DIX Alliance released the second version of Ethernet, Ethernet II. In the 1990s, with the emergence of multi-port bridges, shared Ethernet gradually developed into LAN switches. In 1993, full-duplex Ethernet technology made a breakthrough, with obvious advantages: it could send and receive data at the same time, doubling the speed. 1995 ushered in the golden age of Fast Ethernet development.
In 1998, IEEE released IEEE 802.3z, a 1000Mbps Ethernet standard. In 2002, the 10Gbps Ethernet standard IEEE 802.3ae was officially released. Compared with 1000Mbps Ethernet, 10Gbps only supports full-duplex, and the transmission medium can only be optical fiber. Ethernet is now widely used.
1.2.2Conflict domain _
The conflict domain means that when a device in the physical network transmits data, all other devices on the physical network segment must listen and cannot transmit data. If multiple devices in the same physical network segment transmit data at the same time, an error will occur. Signal conflicts prevent data from being transmitted normally. The typical topology in the conflict domain is a bus structure. All signals are sent on a bus, which is like a one-way street in daily life. All cars run on it, which will cause traffic jams. Data conflicts are shown in Figure 1-4
Figure 1-4 Conflict domain
1.2.3 Broadcast domain
A broadcast domain is to put multiple devices into a group to form a broadcast domain. Any broadcast frame sent by any device in the same broadcast domain will be received by other devices. Just like in a large classroom, the teacher is giving a lecture on the stage, and all students can hear it, but students outside the class cannot hear it; this is because the concept of "class" isolates the broadcast domain. A typical device in the broadcast domain is a switch, which can implement collision domain division. Each interface is a collision domain, but the switch as a whole is still a broadcast domain.
1.2.4 CSMA/CD protocol
The CSMA/CD protocol is a protocol that avoids data signal conflicts in the collision domain. When the host wants to transmit data through the network, because the network line will be monitored and listened to, it first checks whether there is a signal transmitting on the line. If there is no signal being transmitted, the host starts transmitting data and continues to monitor and listen during the transmission process. If other signals are found, the host transmitting data immediately sends a congestion signal. After other hosts detect the congestion signal, they will execute the backoff algorithm and Start a timer that does not transmit any data during the validity period of the timer. In summary, the above can be summarized as "listen at any time, forward when you are free, and wait when you are busy"
The working principle of the CSMA/CD protocol is shown in Figure 1-5.
Figure 1-5 Working principle of CSMA/CD protocol
1.2.5 Half duplex and full duplex
The communication methods include simplex, half-duplex and full-duplex. In simplex communication, the flow of information can only be directed from one party to another at any time; broadcast communication and traditional television systems are simplex communications.
In half-duplex communication, the flow of information can be switched, which means that the roles of data sender and receiver can be changed according to the network environment. However, sending and receiving occupy the same channel, which may cause data conflicts; a common half-duplex communication method is the walkie-talkie system.
In full-duplex communication, the flow of information is arbitrary, regardless of the fixed roles of the sender and receiver; data can be received while sending data. To put it simply, full-duplex communication is like a two-way lane. Sending data and receiving data are separate and do not conflict with each other. The channel utilization rate is also 100%. In daily life, the use of mobile phones to talk is an application of full-duplex communication. A comparison of the three communication methods is shown in Figure 1-6.
Figure 1-6 Simplex, half-duplex, and full-duplex communication methods