We need to start by defining what a standard is. Standards in the computer industry are used to define how the computer hardware and software work together. The standard defines the specifications of how the components are to be designed and assembled so that they work together. Standards assure compatibility so that multiple hardware components will be able to work together. The standard spells out how the data is to be handled at each level as it passes through the computer in its movement to the receiver. In the beginning of networking, these standards were usually proprietary to the manufacturer of the hardware and software. Novell, ARCNET, token ring and Banyan began in this fashion. The user had to have network hardware and network operating system that met the network manufacturer’s specifications in order to use it. Networks between these manufacturers usually did not communicate to each other. The protocols that controlled the data flow were only good for use with the proprietary network operating system. The Ethernet protocols follow this same route.

The IEEE (Institute of Electrical and Electronics Engineers) defines a LAN as:

The Ethernet series standards were developed by IEEE to meet the need of this device communication in local areas. Personal computers that communicate are usually in a local area, a room, a building, an office complex, or for our purposes, two or more computers at home or in a small office. The typical Ethernet LAN is used to share access to data or equipment such as storage devices.  The individual hard disk drive in each computer can be shared so that all the users on the network can access it. Each computer or host on the network can access the files located on another computer depending on the rights and permissions given to the file being used.

LANs also allow users to share devices such as printers. One printer connected to a computer or directly to the network can be used by all the hosts on the network. This is done by sharing the device. Multiple print jobs can be done by that shared printer. A print queue is generated to hold the print jobs is usually established to keep the print jobs in order for printing. There are some devices, printers, scanners, and storage devices, that are “network ready”. Network ready means that the device has a NIC (Network Interface Card) in it to attach directly to the network. Ethernet networks are fully capable of handling shared printing services.

LANs typically are used as message traffic controllers for the users on the network. E-mail from one user to another on the network is provided through electronic mail exchange services. These services include editing and formatting aids such as “Outlook Express” in the Windows OS series. Individual and group addressing capabilities and message notification are usually included. The Internet, which is another network, has e-mail capabilities and there are many e-mail applications such as Eudora, Outlook or Outlook Express can be used here as well. Ethernet networks handle these services just as any other network would.

LANs are used in many ways, one, data transmission that could be sent at low speeds, i.e, small files, or two, very large files that require the LAN high speed bandwidth. One of the basic measures of any network is the speed or bandwidth that it transmits its data. This speed is usually measured as bits per second (bps). As many of us remember, 300 baud or 300 bps using our analog POTS (Plain Old Telephone Service) modems was really fast. Then the speed moved to 9600 bps to 10 Mbps to 100 Mbps as the transmission capabilities of computer networks increased in the past decade. By the way, that 300 baud analog modem was transmitting on a network of sorts. The Ethernet standards call for data rates from 1 to 100 Mbps.

A network consists of several components and software utilities. The basic physical components are computing devices and cabling. The computing devices are known as nodes or hosts. The basic device in the computing component is the Network Interface Card. The NIC is the interface between the computer and the cabling and has the capability of sending and receiving the packets of data that are structured by the Ethernet protocol. Each NIC has an individually assigned address by its manufacturer that is associated with the addressing scheme in the network protocol. This allows each NIC to send and receive data addressed to it. Ethernet follows these standards and protocols.

The cabling is the second basic physical component in the network that must be available for the network to function. The cabling ties the nodes or hosts together and provides the communication medium for data movement. The Ethernet standard works on twisted pair telephone cable, coaxial cable, and fiber optics cable. The NIC being used in the network must be compatible with the type of cable being used. By far, the twisted pair cabling, is being used in the SOHO environment. Coaxial cable has been used in token ring networks and as backbone cabling in large networks. Fiber optics is used extensively for backbone cabling in large networks and for internetworks. The Ethernet standard specifies a maximum cable length of 100 meters, 330 feet. This is more than adequate for most SOHO Ethernet environments.

Most SOHO Ethernet networks use UTP Cat 5 (Unshielded Twisted Pair CATegory 5) communication cable. The Cat 5 cable consists of four twisted pairs of wire, usually solid copper 22 AWG strand insulated wire. The four twisted pairs are enclosed in a protective sheath or jacket. Cat 5 UTP cable is easy to install and terminate. Cable termination is very important. Each cable end must be terminated in a communication plug. These plugs look like a telephone wire plug, but are larger to accommodate eight wires instead of four. The name for these plugs is RJ 45. The eight individual wires are inserted into the plug in a specific order to insure communication. The Cat 5 UTP cable and RJ45 plugs can be purchased in bulk so that the cabling installation can be custom to the network, or ready made Cat 5 UTP cable can be purchased in the lengths that closely fit the network. Of course, the NICs installed in the computers must be able to receive the RJ 45 cable plugs.

Most NICs for Ethernet networks use the RJ 45 jacks that receive the RJ 45 plugs. In the recent past, many Ethernet NICs had both RJ 45 jacks and BNC connectors for coaxial cable. The SOHO network has nearly completely gotten away from using coaxial cable.

The network topology, the layout technique, must be considered in the Ethernet SOHO environment. The topology determines how the computers will be connected. Ethernet relies on a Star configuration. Each node in the network connects to a central hub. This hub is a physical device. It usually requires electrical power. It has RJ 45 jacks to receive each of the cables from the nodes in the network. The jacks or ports as they are called, are in multiples of two, usually four, eight, sixteen or larger, depending on the number of nodes in the network. The hub acts as a passive communication traffic manager, receiving and sending the network communication to all other nodes. A smart version of the Ethernet hub, called a switch, does understand the addressing scheme in the network and can route network traffic to the proper address.

Other specialized network devices can be used in the Ethernet network to make it function more efficiently. Routers are one example of these devices. Ethernet routers can make the data sending and receiving process function easier.

Each computer in the network functions in accordance with the role assigned to it. Some networks act as peer-to-peer communication devices. In peer-to-peer networks, each node is equal to all other nodes and must keep track of every other computer to keep the addressing scheme straight. This is an acceptable way to use the network as long as the nodes are ten or less. In most serious networks, the best way to set up its functioning is to make at least one of the computers function as a server. Servers do just that, they provide services such as e-mail exchange, printing, and data storage to other computers called clients. The Ethernet standard handles these arrangements.

The network must have a ‘network ready’ operating system to complete the communication processing between the nodes in the network. Within this OS are communication protocols that handle the data flow within the computer and in its transmission process. The protocol most commonly used today is the TCP/IP (Transmission Control Protocol/ Internet Protocol). TCP/IP processes within the OS and is hardware independent. It transmits over Ethernet. TCP/IP handles the addressing and packet assembly of data that is to be transmitted on the sending computer and receives the data packets on the receiving computer. Windows 9X series, Windows NT/2000/XP, Apple, UNIX, and Linux, to name a few network ready OSs can use TCP/IP over Ethernet networks.

Summary
The Ethernet standard has become the network specification of choice for SOHO networks because of its ease of understanding and implementation. The standard is well tested and contains very few problems. It is easily deployed in the SOHO network. Vendors have provided a very good mix of hardware that is full capable of operating in the SOHO network. The Ethernet components are stock items at computer stores. The vendors have done a relatively good job of providing working drivers to set the components such as Ethernet NICs in today’s Windows and other OSs. The other half of the Ethernet network, cabling, is also readily available at computer stores as shelf merchandise.

Ethernet works over wireless networks as well. The NICs and Access Points for the wireless networks usually plug into Ethernet RJ 45 cable connections for routers and modems. I have a wireless Access Point in my home network.

The actual detailed setup may have some NIC driver problems, or the router may require some special setup, but, for the most part, an Ethernet network can easily be setup in the SOHO environment.