An early version of Xerox Ethernet was released in 1975 and was designed to connect 100 computers at a speed of 2.94 megabits per second over a kilometer-long cable. The design became so successful at that time that Xerox, Intel and Digital Equipment Corporation (DEC) issued the 10Mbps Ethernet standard that is widely used in computer networks today. In addition, the standard Ethernet speed is 100Mbps, known as Fast Ethernet.
Ethernet originated from a WAN development at the University of Hawaii in the late 1960s known as “ALOHA”. The university has a large campus geographical area and is willing to connect computers spread across the campus into a campus computer network.
The process of standardizing Ethernet technology was finally approved in 1985 by the Institute of Electrical and Electronics Engineers (IEEE), with a standard known as Project 802. The IEEE standard was subsequently adopted by the International Organization for Standardization (ISO), making it an international and worldwide standard intended to form a computer network. Because of its simplicity and reliability, Ethernet has survived to this day, and is even the most widely used network architecture.
- Types of Ethernet
When viewed from the speed, Ethernet is divided into four types, namely:
- 10 Mbit / sec, which is often referred to as Ethernet only (standard used: 10Base2, 10Base5, 10BaseT, 10BaseF)
- 100 Mbit / sec, often referred to as Fast Ethernet (standard used: 100BaseFX, 100BaseT, 100BaseT4, 100BaseTX)
- 1000 Mbit / sec or 1 Gbit / sec, often referred to as Gigabit Ethernet (standard used: 1000BaseCX, 1000BaseLX, 1000BaseSX, 1000BaseT).
- 10000 Mbit / sec or 10 Gbit / sec. This standard has not been widely implemented.
|10 Mbit/second||10Base2, 10Base5, 10BaseF, 10BaseT||IEEE 802.3||Ethernet|
|100 Mbit/second||100BaseFX, 100BaseT, 100BaseT4, 100BaseTX||IEEE 802.3u||Fast Ethernet|
|1000 Mbit/second||1000BaseCX, 1000BaseLX, 1000BaseSX, 1000BaseT||IEEE 802.3z||Gigabit Ethernet|
The Ethernet specification defines the functions that occur at the physical layer and the data-link layer in the OSI seven-layer network reference model, and how to package data into a frame before it is transmitted on a cable.
Ethernet is a network technology that uses the Baseband transmission method which sends its signal serially 1 bit at a time. Ethernet operates in half-duplex mode, which means that each station can receive or send data but cannot do both at the same time. Fast Ethernet and Gigabit Ethernet can work in full-duplex or half-duplex mode.
Ethernet uses the Carrier Sense Multiple Access with Collision Detection media access control method to determine which stations can transmit data at any given time through the media used. In a network that uses Ethernet technology, each computer will “hear” first before “talking”, meaning that they will see the condition of the network if no other computer is transmitting data. If there is no computer transmitting data, then every computer that wants to send data can try to take over the network to transmit the signal. So, it can be said that a network that uses Ethernet technology is a network made on the basis of First-Come, First-Served, rather than delegating control signals to the Master Station as in other network technologies.
If two stations try to transmit data at the same time, there will probably be a collision, which will cause the two stations to stop transmitting data, before finally trying to send it again at random time intervals (measured in units millisecond). The more stations in an Ethernet network, the greater the number of collisions and the poor network performance. Ethernet performance should be 10 Mbit / sec, if in the network installed 100 nodes, generally only produces performance ranging from 40% to 55% of the expected bandwidth (10 Mbit / sec). One way to deal with this problem is to use an Ethernet switch to segment the Ethernet network into multiple collision domains.
- Ethernet frame
Ethernet transmits data through network cables in the form of data packets called Ethernet Frames. An Ethernet frame has a minimum size of 64 bytes, and a maximum of 1518 bytes, with 18 bytes used as information about the source address, destination address, network protocol used, and some other information stored in headers and trailers (footers). In other words, the maximum amount of data that can be transmitted (payload) in a single frame is 1500 bytes.
Ethernet uses several methods to encapsulate data packets into Ethernet frames as:
- Ethernet II (used for TCP / IP)
- Ethernet 802.3 (also known as Raw 802.3 in Novell network systems, and used to communicate with Novell NetWare version 3.11 or earlier)
- Ethernet 802.2 (also known as Ethernet 802.3 / 802.2 without Subnetwork Access Protocol, and used for connectivity with Novell NetWare 3.12 and later)
- Ethernet SNAP (also known as Ethernet 802.3 / 802.2 with SNAP, and made as compatibility with Macintosh systems running TCP / IP)
Unfortunately, each of the Ethernet frame formats above is incompatible / compatible with one another, making it difficult to install heterogeneous networks. To overcome this, configure the protocol used via the operating system.
Ethernet can use any physical network topology (can be a bus topology, ring topology, star topology or mesh topology) as well as the type of cable used (can be a coaxial cable (can be Thicknet or Thinnet), copper cable (UTP cable or STP cable), or fiber optic cable). However, star topology is preferred. Logically, all Ethernet networks use the bus topology, so one node will put a signal on the bus and the signal will flow to all other nodes that are connected to the bus.
- Ethernet Operation Mode
The two main modes of operation of Ethernet are full duplex and half duplex.
- Full Duplex
In full-duplex communication, two communicating parties send information and receive information at the same time, and generally require two lines of communication.
Full-duplex communication can also be achieved using multiplexing techniques, where signals that travel in different directions will be placed in different time slots. The weakness of this technique is that this technique cuts the transmission speed which may be in half.
- Half Duplex
Half-duplex is a communication mode where data can be transmitted or received in both directions but cannot be shared. The simplest example is a walkie-talkie, where two users must press a button to speak and release the button to listen. When two people use a walkie-talkie to communicate at one particular time, only one of them can talk while the other party listens. If both try to talk together, the “collision” condition occurs and both walkie-talkie users cannot listen to what they are sending.
The only difference between them is that a half duplex connection allows data traffic to flow in both directions, but not simultaneously. Whereas full duplex allows sending and receiving data at the same time, effectively increasing the transmission rate doubled.
There is another mode of operation, simplex, which only allows one-way data transmission. Simplex is a form of communication between two parties, where signals are sent in one direction. This transmission method is different from the full-duplex method which is able to send signals and receive them simultaneously at one time, or half-duplex that is able to send signals and receive signals even if not at one time. Simplex transmission occurs in several communication technologies, such as television broadcasts or radio broadcasts.
Simplex transmission is not used in network communications because nodes in a network generally require two-way communication. Indeed, some communication in the network, such as video streaming, looks like a simplex, but actually the communication traffic occurs in two directions, especially if the TCP protocol is used as the transport layer protocol.
However, this mode of operation is not used on Ethernet. As a formal specification, 10BaseT and 100BaseTX support full duplex, but in practice this capability is only implemented on 100BaseTX.