BNC (Bayonet-Neill-Concelman)

BNC is an abbreviation for Bayonet-Neill-Concelman. The B indicates that the connector is a bayonet-type connection, and Neill and Concelman are the inventors of the connector. You may also hear this connector called a British Naval Connector.

10Base-2 Ethernet

Though not as common as it once was, 10Base-2 is still an excellent way to connect a small number of computers together in a small physical area such as a home office, classroom, or lab. The 10Base-2 Ethernet uses thin coaxial (RG-58/U or RG-58 A/U) to connect computers together. This thin coaxial cable is also called thinnet. Coaxial cable and network-interface cards use a special connector called a BNC connector. On this type of connector, the male is inserted into the female, and then the male connector is twisted 90 degrees to lock it into place. A BNC T-connector allows two cables to be connected on each side of it, and the middle of the T-connector plugs into the network-interface card. The thinnet cable never connects directly to the network-interface card. This arrangement is shown in Figure

10Base-F Ethernet

Specifications for using Ethernet over fiber-optic cable existed back in the early 1980s. Originally, fiber-optic cable was simply used to connect repeaters whose separation exceeded the distance limitations of thicknet cable. The original specification was called Fiber Optic Inter Repeater Link (FOIRL), which described linking two repeaters together with fiber-optic cable up to 1,000 meters (3,280 feet) in length.

10Base-T Ethernet

For over 10 years, 10Base-T (the T stands for twisted pair) Ethernet reigned as king of the network architectures. There is a good reason for this: 10Base-T Ethernet will work over any regular Category 3 or better UTP cabling, and UTP cabling is cheap to install, reliable, and easy to manage.

Ethernet

Ethernet is the most mature and common of the network architectures. According to technology analysts IDC (International Data Corporation), Ethernet is used in over 80 percent of all network installations. In some form, Ethernet has been around for over 30 years. A predecessor to Ethernet was developed by the University of Hawaii (called, appropriately, the Alohanet) to connect geographically dispersed computers. This radio-based network operated at 9,600Kbps and used an access method called CSMA/CD (Carrier Sense Multiple Access/Collision Detection), in which computers "listened" to the cable and transmitted data if there was no traffic. If two computers transmitted data at exactly the same time, the nodes needed to detect a collision and retransmit the data. Extremely busy CSMA/CD-based networks became very slow when collisions were excessive. In the early 1970s, Robert Metcalfe and David Boggs, scientists at Xerox's Palo Alto Research Center (PARC), developed a cabling and signaling scheme that used CSMA/CD and was loosely based on the Alohanet. This early version of Ethernet used coaxial cable and operated at 2.94Mbps. Even early on, Ethernet was so successful that Xerox (along with Digital Equipment Corporation and Intel) updated it to support 10Mbps. Ethernet was the basis for the IEEE 802.3 specification for CSMA/CD networks.
Over the past 25 years, despite stiff competition from more modern network architectures, Ethernet has flourished. In the past 10 years alone, Ethernet has been updated to support speeds of 100Mbps and 1000Mbps; currently 10 Gigabit Ethernet is being deployed over optical fiber and research is progressing to make it available over UTP. Ethernet has evolved to the point that it can be used on a number of different cabling systems.

Network Architectures

The ANSI/TIA/EIA-568-B cabling Standard covers almost any possible combination of cable necessary to take advantage of the current network architectures found in the business environment.  These network architectures include Ethernet, Token Ring, Fiber Distributed Data Interface (FDDI), Asynchronous Transfer Mode (ATM), and 100VG-AnyLAN. Although the predominant cabling infrastructure is UTP, many of these architectures are capable of operating on other media as well. Understanding the different types of cable that these architectures utilize is important.

Ring Topology

A ring topology requires that all computers be connected in a contiguous circle, as shown in Figure  The ring has no ends or hub. Each computer in the ring receives signals (data) from its neighbor, repeats the signal, and passes it along to the next node in the ring. Because the signal has to pass through each computer on the ring, a single node or cable failure can take the entire ring down.

A true ring topology is a pain in the neck to install cable for because the circular nature of the ring makes it difficult to expand a ring over a large physical area. Token Ring is a ring topology. Even though Token Ring stations may be connected to a central MAU (and thus appear to be a star topology), the data on the Token Ring travels from one node to another. It passes though the MAU each time.