Ethernet is a collection of various computer networking technologies that are meant for local area networks or LANs. The internationally accepted standard for Ethernet is IEEE 802.3. It was the most used networking technology in the 1990s and replaced coaxial cable, token ring, ARCNET etc.
The basic idea behind Ethernet was to help computers communicate over a coaxial cable such that it could work as a broadcast transmission medium. Ethernet enabled its administrator to easily detect any collision that may exist on the network as against a radio broadcast system. The use of twisted-pair cable has made Ethernet the most widely used networking technology of today for connecting Local Area Networks or LANs.
- CSMA/CD shared medium Ethernet
This type of Ethernet makes use of a shared coaxial cable that is wound around a building or campus. The technique called Carrier Sense Multiple Access with Collision Detection (CSMA/CD) controls the way that the individual computer systems share the channel. This is seen as being simpler than token ring or token bus technologies.
- Ethernet repeaters and hubs
Ethernet connections have a problem of signal degradation. This becomes a major consideration when data is to be transferred over a long distance since the maximum size that Ethernet segments supported was around 500 meters. To solve this problem and enable the use of larger networks, a repeater was placed between two cables. This repeater takes signals from one cable and repeats it into another cable. However, installing repeaters which had only single cable ports were found to be uneconomical. As the star topology became more and more important, a need was felt to have a repeater that supported the use of more than one port. Thus, multi-port repeaters were introduced which came to be known as hubs. Thus, many cables could be connected through the same repeater or hub.
Despite the physical star topology, hubbed Ethernet networks use half-duplex and CSMA/CD, with only minimal activity by the hub, primarily the Collision Enforcement signal, in dealing with packet collisions. Every packet is sent to every port on the hub, so bandwidth and security problems aren't addressed. The total throughput of the hub is limited to that of a single link and all links must operate at the same speed.
- Bridging and switching
Bridging was developed as a technique to deal with a characteristic problem of Ethernet hubs. The problem with hubs was that they relayed the traffic to all nodes on the network instead of concentrating the traffic to its intended destination hence limiting the number of nodes that could be connected to the network. Bridging isolated the physical layer by working on the data link layer only and only ‘healthy’ packets were sent on to the Ethernet segments. This way, any errors that were caused by collision of traffic or the packets were isolated. Ethernet bridges are capable of detecting where the devices exist through the MAC addresses and if the destination device is not located in a particular direction, the packets were not sent in that direction.
While the bridges work on the data link layer, switches perform the same function as bridges but on the physical layer or rather, at a hardware level. Thus, packets can be forwarded at the full speed of the hardware without delaying them as a result of bridging.
Both, bridges and switches, can work with equipment that has various speeds, thus allowing the use of such equipment in the same network. Also, the extra processing required for detecting collisions of traffic over the network is not required to be done by the nodes on the network as the bridges and switches take care of
