14 DAY TRIAL // An important aspect for modern PCs is the possibility of connecting them to the Internet or to a network made up of many other computers. After all, the Internet can be considered to be the widest network that covers almost the entire planet. Connecting to the Internet through a local area network (LAN) has become a more popular choice when compared to modems and this is ultimately determined by the need for improved transfer speeds. LANs themselves are based on a concept that is known as the Ethernet. The main idea of LANs and Ethernet standards is to facilitate high speed data transfers between numerous PCs.
The web-like technology Ethernet mainly defines an array of frame-based computer networking technologies for local area networks. The standardized denomination for the Ethernet is IEEE 802.3. It describes star-topologies and twisted pair wiring forms, which contributed the most to the expansion of LANs during the 1990s and all the way up to now. The popularity of the Ethernet standard largely replaced competing LAN standards such as coaxial cable Ethernet, token ring, FDDI (fiber distributed data interface), and ARCNET (Attached Resource Computer NETwork). The beginning of the new century brought Wi-Fi and wireless LAN standards starting with IEEE 802.11, and the present trend is to consolidate the older Ethernet structures with new wireless ones.
The Ethernet LAN structures consist of two main elements: -the interconnecting media; -the network nodes.
The interconnecting media allows data signals to propagate between the nodes. With the first Ethernet networks, the interconnecting media was represented by coaxial cable with a single inner connector. Nowadays, either Unshielded Twisted Pair (UTP) or a Shielded Twisted Pair (STP) cables is normally used. Higher data rate systems may also use superior fiber optic cables.
Furthermore, the network nodes split into two categories:
1. Data Terminal Equipment (DTE) - these devices are either the source or destination of the data being sent. We usually include PCs, file servers, print servers and other derived forms in this category. 2. Data Communications Equipment (DCE) - this category includes devices that may receive and forward the data frames across the network. They are also known as intermediate nodes. Such devices include repeaters, routers, switches or even modem-based communication interface units.
The Ethernet standard terminology also includes the term "frame." The nodes communicate in short messages called frames, which are variably sized chunks of information.
Frames are similar to sentences in human language. For each human language, people have rules for constructing sentences. It is universally known that each sentence must contain a subject and a predicate (though there are exceptions, as always). Analogically, Ethernet protocols specify a set of rules for constructing frames. There are explicit minimum and maximum lengths for frames, and a set of required pieces of information that must always appear in the frame. Each frame must include, for example, both a destination address and a source address, which identify the recipient and the sender of the message. The address uniquely identifies each node, just as a name identifies a particular person. For this reason, Ethernet-based devices should never have the same address.
Ethernet-based communication systems developed three main types of topological configurations. According to actual requirements, these models are:
a) Point to point (P2P) - this is the simplest configuration as only two network nodes are used. It may be a DTE to DTE, DTE to DCE, or even a DCE to DCE. In this case, the cable is known to be a network link. Links like these are used to transport data from one node to another.
b) Coaxial bus - a rarely used type these days. The systems use a coaxial cable where the network nodes are located along the length of the cable. The segment lengths are limited to a maximum of 500 meters, and it is possible to place up to 1024 DTEs along its length. Official support for this model has been discontinued although there may still exist legacy systems using it.
c) Star network - it is the dominant topological model since the early 1990s. It consists of a central network node known as a multiport repeater or hub, or a network switch. All the connections to other nodes radiate out starting from this device via P2P configurations.
The experimental Ethernet system described in that paper ran at an impressive speed (for that period) of 3 Mbit/s, and had 8-bit destination and source address fields. These first Ethernet addresses were different from today's global addresses.
In 1979, Metcalfe left Xerox to promote his invention and to extend it to the evolving PC market. He thus founded the 3Com company and managed to convinced DEC Computers, Intel, and Xerox to work together to promote Ethernet as a network standard. This was to be known as the "DIX" standard, for "Digital/Intel/Xerox," featuring speeds of 10 megabits/second, with 48-bit destination and source addresses and a global 16-bit type field. The standard was first issued on September 30, 1980. It quickly became a serious competitor for the existing two largely proprietary systems - token ring and ARCNET. In the process of dethroning the latter standards, 3Com resolved to become a major company.
Computer manufacturers quickly adopted the Ethernet as a standard feature for their PC models, and this allowed 3Com to build a business around selling add-in Ethernet network cards. Since then, Ethernet products have continually enjoyed speed per cost advantages over other LAN implementations. Even the recent Wi-Fi technologies use Ethernet for connecting to the rest of the network.
Ethernet Evolution As stated before, Ethernet was originally envisaged as a communication means that connected a number of nodes over a shared coaxial cable acting as a broadcast transmission medium. The incipient Ethernet stages bore similarities to radio system, although there are major differences, such as the fact that it is much easier to detect collisions in a cable broadcast system than a radio broadcast. The coaxial cable design was directly likened to the ether and thus, the whole deal became to be known as "Ethernet".
The coaxial cable was later replaced with point-to-point links and the additional hubs and/or switches emerged in order to reduce installation costs, increase reliability, and enable point-to-point management and troubleshooting. StarLAN was the first step in the evolution of Ethernet from a coaxial cable bus to a hub-managed, twisted pair network. The introduction of twisted-pair wires propelled the Ethernet technology to a large-scale commercial success.
Despite the huge changes in Ethernet from a thick coaxial cable bus running at 10 Mbit/s to point-to-point links running at 1 Gbit/s and beyond, all post-experimental Ethernet generations share the same frame formats (and hence the same interface) and can be readily interconnected due to backward compatibility.
The wide adoption of Ethernet technologies, as well as the ever-decreasing cost of the hardware needed to support them coupled with the reduced panel space needed by more recent twisted pair Ethernet designs, led to the integration of Ethernet cards directly into PC motherboards, removing the need for installation of a separate network card. The present trend is to include two Gigabit integrated network interface chipsets for mainstream and high-end motherboards.
The limits of Ethernet technology There are practical limits to the size of an Ethernet network. A primary concern is the length of the shared cable. Electrical signals propagate along a cable with near-light speeds, but their speeds decrease as they travel, and in addition, electrical interference from neighboring devices can scramble the signal. So, the shorter the cable is, the clearer and faster devices at opposite ends can receive each other's signals. This places a distance limitation on the maximum separation (network diameter) between two devices connected to an Ethernet-based configuration. Additionally, there are practical limitations to the number of devices that can simultaneously connect to a single network. If too many devices are attached to the network, every device may have to wait a variable amount of time before getting a chance to transmit. These problems appear to have been solved with the introduction of special devices which aren't, however, specific to Ethernet standards, but can be adapted to them, playing roles in other network technologies as well.
We stop here for the moment. The next article will present a more in-depth look at how Ethernet networks work.