Basic Protocol Ethernet

During the 90s of last century, many protocols exist in different network architectures. Engineers have developed different types of protocols, each for a specific data communication such as using data, voice, broadcasting, data centers, and so on. Among which one can find Token Ring, FDDI, ATM and more.

However, because of the simplicity and low cost of Ethernet components, won in the end, and is now the most widely used protocol for Local Area Network (LAN). Faster Ethernet protocol has been defined to meet many of the above applications, and network.

In fact, Ethernet is a family of protocols, each with a different speed and they are all based on the transfer of the packet, as defined by IEEE 802.3 standard group. Today, there are Ethernet protocols and components ranging from 10Mbit/sec (10 million bits per second) at 100Mbit/sec (also known as Fast Ethernet), 1 Gbit / sec (1 billion bits per second) , 10Gbit/sec, and lately 100Gbit/sec 40Gbit/sec.

Ethernet is a physical layer protocol (Layer 1 and 2 of the OSI 7 layers). Is defined as an Ethernet frame (packet) is transferred to the cable or fiber, and how Ethernet switches in the path can direct the course of their destiny.

Ethernet can run on twisted pair copper and fiber cables. Copper cables can usually transfer Ethernet packets of up to 100 meters, and are mainly used in building internal networks. Fiber cables can reach tens of kilometers and are used to link between the campus and outside the cities.

Is a physical layer protocol, Ethernet is a data for upper layer protocols like IP (Internet Protocol). In fact, most cases where people discuss the IP protocol, is supposed to discuss the Ethernet protocol for support. The IP protocol used for communication between two logically separate networks. That is, Ethernet enables direct communication in a logical network, called VLAN (Virtual LAN), while it should use the IP for the proper routing of Ethernet packets between VLANs.

Most electronic devices we use, such as personal computers and laptops, to communicate by sending each Ethernet packet. They noted that the Ethernet stations. Each station has a unique address (a unique number consists of 48 bits). This is called a MAC address. Each station has a network interface card (NIC) that is aware of this issue and do not leave a package with other Ethernet MAC address to enter the station. Furthermore, this MAC address is added to a packet sent by the NIC. Finally, the package is the only source direction along its path.

The main component is an Ethernet network with an Ethernet switch (the sake of clarity we are talking about here on switched Ethernet, and ignored older system shared Ethernet, which is almost obsolete). Switch is connected to the remote or copper cable or fiber optic cables and connect the different levels of variable speed drives. Since all units, no matter how fast you use the same packet, the switch allows all stations, regardless of speed for the connection.

As defined in the IEEE 802.3 group, the Ethernet switch is designed to detect the source of an Ethernet packet (Ethernet Source Address) and the packet destination (Ethernet destination address). The source and destination addresses are the first parameters of the Ethernet packet, and reads the switch. The switch address table continually learn and keep track of addresses contained in the packets arrive in time to move and organize the address table so that each address with a port, some of which became the source address of the switch. This table is available for every incoming packet to the output port directly relevant, so that the packet should reach its destination at the end.

Address detection, management learning and decision to approve the package, all done by physical means, as soon as possible. This is why Ethernet can reach 100Gbits/second.