Ethernet

Ethernet emerged from a project of the companies Digital Equipment, Intel and Xerox in the seventies, which became known as DIX. This project aimed at sharing a transmission medium between several data stations with equal rights and was designed for the local area. The structure of this local area network was that of a bus to which all data stations could be connected. The bus concept was limited in its extent, in the number of stations that could be connected and in data rate to 3 Mbit/s and was therefore only suitable for the local area. The DIX project became Ethernet in 1982. The specification and standardization was taken over by the IEEE, which published the specifications DIX Ethernet V2.0. Since Ethernet developed rapidly in the following decades and the name Ethernet is also used for high-speed variants, the original Ethernet is also referred to as such. The classic Ethernet The Ethernet specified by the IEEE working group 802.3 is a local network to whose transmission medium equal data stations are connected. The data stations work with a stochastic access method in which they detect signals on the transmission medium and then, if no other signal is present, use it for their own data transmission. The access method is CSMA/CD and can lead to collisions between the transmitted data signals. The original Ethernet according to 10Base-5 operates with a transmission speed of 10 Mbit/s. All data stations have equal rights and are connected directly to the transmission medium. The number of data stations is limited to max. 256 stations per Ethernet segment, the length expansion is limited by the signal runtime and the frame length. As the segment length increases, the propagation time between the two stations furthest apart increases, which has a negative effect on the time required to detect a collision. All three parameters, the number of stations, the frame length and the segment length, determine the performance of Ethernet. The shorter the frame length and the higher the number of stations ready to transmit, the lower the performance. The frame length of the Ethernet frame affects performance in that the ratio of user data to frame length decreases as the frame size becomes smaller. The permissible network expansion of an Ethernet depends on the Ethernet variant and thus on the transmission speed on the network. In general, the expansion decreases with increasing transmission speed. If we consider the classic Ethernet with 10 Mbit/s transmission speed, the permissible network expansion is determined from the round-trip delay on the transmission medium. This is determined by the propagation time between the two most distant stations. With a cable-specific signal propagation delay of 0.77c, this results in a propagation delay of 2.165 µs for a 500 m segment. Based on the minimum packet length of an Ethernet frame of 64 bytes and the transmission rate of 10 Mbit/s, the round trip time between the two most distant stations is 51.2 µs, so the simple propagation delay is 25.6 µs. This round trip delay corresponds to an extension of about 5 km. Taking into account the signal propagation times in the transceiver cables and the delay times of the active components, Ethernet extensions of about 3,000 m are thus possible. Ethernet layer model for classic Ethernet

Englisch:	Ethernet
Updated at:	29.09.2019
#Words:	535
Links:	project, Digital Equipment, Intel, Xerox (Ethernet) (DIX), transmission medium, data, area
Translations:	DE
Sharing: