Overview of Networking Technologies	3.2/5.0 (6 votes total) Rate: Select 1 - Terrible 2 - Poor 3 - Average 4 - Good 5 - Outstanding
Team uCertify October 05, 2006

A lot of processes are required for running a network in addition to the network media through which the connections are made. To ensure that all these processes operate flawlessly in complex networking environments, there are some common standards and technologies that are used for different types of networks.

The Institute of Electrical and Electronic Engineers (IEEE) develops these standards and technologies, which are followed by network engineers throughout the world to design different types of networks. The institute set up a task force known as the 802 committee in 1980 to develop standards for establishing communication between different computers in a network. These standards ensure the flow of data in the data link layer of the OSI model. The most common data link protocols directed by these standards and used in the present day networks are Ethernet (IEEE 802.3), Token Ring (IEEE 802.5), and Wireless LAN (IEEE 802.11b). There is one more popular LAN protocol known as FDDI, which is developed by American National Standards Institute (ANSI).

Ethernet (IEEE 802.3)

Conceived in the 1960s, the Ethernet is the oldest and most popular data link layer protocol (or network technology) used in today's networks. Ethernet networks use a bus or star topology and control the flow of data through the media access control (MAC) method, known as Carrier Sense Multiple Access Collision Detection (CSMA/CD). The use of CSMA/CD ensures that each computer in the network can send its signals over the network. To send signals over the network, a computer waits for the network to be free of any traffic; and if the network is free, it sends its signals, which travel through the network and are received by the destination computer. Sometimes, more than one computer sends its signals over the network, which results in collision. Collisions in these types of networks cannot be avoided, as CSMA/CD can detect them only when they occur. It then sends the data over the network again to compensate the data loss.

Ethernet networks run at the speeds of 10 to 1,000 Mbps, depending on the type of topology and cabling used. Ethernet technology is widely implemented in Star topology using coaxial or fiber optic cables and in bus topology using UTP cable.

Types of Ethernet

• 10 Base5
  
  These networks are also known as Thick Ethernet, as they use a thick coaxial cable called RG 8 for data transmission. The number 10 specifies that it provides a transmission speed of 10 Mbps for the network. The number 5 denotes that the maximum segment length provided by these networks is 500 meters. The maximum length of these networks can be increased up to 2.5 kilometers (1.5 mile) by using repeaters. 10 Base5 networks use bus topology, and each computer is attached to the network by using separate cables called Attachment Unit Interface (AUI). The important specifications of these networks are stated in the table below:
  

• 10 Base2
  
  The common name used for these types of networks is Thin Ethernet as they use a thin coaxial cable (RG 58), which is inexpensive, flexible and easy to handle. 10 Base2 networks also use bus topology and provide a speed of 10 Mbps with a maximum segment length of 185 meters. These networks do not require an AUI, and the cables are connected directly to the network interface cards of computers. The specifications of 10 Base2 networks are as follows:
  

• 10 Base-T
  
  These networks use unshielded twisted pair cable that is more flexible, cost effective, and easier to install than the coaxial cable. These networks use star topology implemented logically as a bus in which a separate cable to a hub connects every computer in the network. A 10 Base-T network uses category 3, 4 or 5 cables with a maximum cable segment length of 100 meters. These networks are cost effective and easy to troubleshoot and provide a speed of 10 Mbps. 10 Base-T networks have the following specifications:
  

• 100 Base-TX
  
  The number 100 used in the name indicates that these networks run at the speed of 100 Mbps using category 5 UTP cable. The topology used in these networks is star. The maximum cable segment length of 10 Base TX is 100 meters. These networks are also known as fast Ethernet and are the most widely used networks today. 100 Base TX Ethernet uses two of the four wire-pairs in the category 5 UTP cable. The specifications of these networks are given below:
  

• 100 Base-T4
  
  These networks also run at the speed of 100 Mbps but use category 3 UTP cable. 100 Base T4 networks use star topology, and the maximum segment length of the cable is 100 meters. These networks use all the four -wire- pairs of the category 3 cables, in which two pairs transmit and receive data, and the other two carry the data in both directions to provide high speed. These networks have the following specifications:
• 100 Base-FX
  
  This is also a fast Ethernet network that runs at the speed of 100 Mbps using Fiber optic cable with the segment lengths of up to 412 meters for half duplex connections and 2,000 meters in full duplex connections. The similarity between 100 Base FX and other fast Ethernet specifications is the use of star topology. The following table lists the 100 Base FX specifications:
  

• 1000 Base-T
  
  It is a Gigabit Ethernet specification, which runs at a speed of 1000 Mbps and has the maximum cable segment length of 100 meters. Although 1000 Base T requires category 6 UTP cable, it can also run on a category 5e (enhanced) cable standard. These networks use all the four wire pairs of the UTP cable for a faster flow of data in both directions simultaneously. Computers in a 1000 Base T networks are connected using a star topology. The main specifications of this standard are given below:
  

• 1000 Base-CX
  
  This specification was in use before the launch of 1000 Base-T. It was used for the networks spread over short distances, but it has now become obsolete. These networks used shielded twisted pair (STP) cables, and the maximum cable segment length was 25 meters.
  
• 1000 Base-SX
  
  This Ethernet specification utilizes the capabilities of fiber optic cables to provide a speed of 1000 Mbps. SX stands for short wavelength laser as it uses short wavelength laser over multimode fiber. The maximum segment length for 1000 Base-SX is 550 meters.
  
• 1000 Base-LX
  
  Like 1000 Base-SX, this specification is also a fiber optic variant of 1000 Base-CX, but it uses long wavelength laser over multimode and single mode fiber to run at the speed of 1000 Mbps. The maximum segment length of these networks is also 550 meters.
  
• 10GBase-SR
  
  The 10 Gigabit Ethernet is the latest Ethernet standard approved by the IEEE in June 2002. It uses only fiber optic cable and operates in full duplex mode, providing the speed of 10 Gbps. These standards provide the flexibility for use in LANs, MANs and WANs.
  
  The 10Gbase-SR operates on short wavelength (850 nm) multimode fiber and has a maximum segment length of 2 meters to 300 meters depending on the type and quality of the multimode fiber.
  
• 10GBase-LR
  
  The 10Gbase-LR Ethernet specification operates on long wavelength (1310 nm) single mode fiber. It provides a maximum segment length of 2 meters to 10 kilometers (or more), depending on the quality and type of the single mode fiber cable.
  
• 10GBase-ER
  
  This is also a 10 GB Ethernet standard, which operates on extra long wavelength (1550 nm) single mode fiber. The maximum segment length for this specification is from 2 meters to 40 kilometers, depending on the quality and type of cable used.
  

Token Ring (IEEE 802.5)

Originally developed by IBM, the token ring is an intricate but highly dependable networking technology that follows the IEEE 802.5 standard. The type of topology used in this technology is physically a star, but implemented logically as a ring, in which all the computers are attached to a central unit called a multistation access unit (MAU OR MSAU). Token ring networks use token passing to send their signals over the network. Token is a type of data packet, which circulates in the entire network. If the token is free, the computer waiting to send data takes it, attaches the data and the destination address to the token, and sends it. When the token reaches its destination computer, the data is received. Then, the token gets back to the originator. If the originator finds that the message has been received, it removes the message from the token. Now, the token is free and can be used by the other computers in the network to send data.

Token ring networks are more fault-tolerant than the Ethernet, as the MSAU ensures that the failure of a single computer does not bring the entire network down. It is an intelligent device which can identify the failing computer in the network, and then it bypasses it to correct the errors.

The modern day token ring networks use unshielded twisted pair (UTP) cable and run at the speed of 16 Mbps as opposed to the original token ring networks developed by IBM that used shielded twisted pair (STP) cable and ran at 4 or 16 Mbps.

Wireless LAN (IEEE 802.11b)

The IEEE 802.11b standard applies to wireless LAN Networks that use radio waves as the transmission medium. These networks have their separate media access control mechanism known as Carrier Sense Multiple Access with Collision Avoidance (CSMS/CA), that works like the CSMA/CD mechanism used by the Ethernet. The difference is that in this protocol, when a computer sends its signals, the receiving computer performs a cyclical redundancy check and sends an acknowledgement message to the sending computer if no error is detected.

As these networks are wireless, they do not use the traditional topologies used by other networks. There are two types of wireless networks, namely Ad hoc network and Infrastructure network. In an Ad hoc network, a group of computers are connected using the IEE 802.11b data link protocol and communicate as peers. In an Infrastructure network, the computers are connected using wireless access points that are connected to cables.

The wireless LANs run at the transmission speed of 11 Mbps with a fallback to 5.5, 2, and 1 Mbps in the 2.4 GHz bandwidth. The radio transmission medium used by these networks is called Direct Sequence Spread Spectrum (DSSS).

FDDI (Fiber Distributed Data Interface)

Developed by the American National Standards Institute, FDDI is a ring-based network that uses fiber optic cables to provide very fast and reliable communication between the connected computers. It uses token passing to control the network access, but unlike the token ring networks, it does not use a hub; instead, it uses a central device called concentrator to connect the computers in the network. In these networks the computers are connected using physical ring topology.

There are two types of configurations used by FDDI networks, namely class A and class B configurations. In class A, a double ring topology is used in which the computers are connected to two rings. The signals travel in opposite directions on both the rings, and if there is a fault in one ring, the receiving computer can still receive the signal through the other ring. These networks provide a better fault tolerance. Class B networks use a single physical ring and is, therefore, less fault tolerant.

FDDI networks run at the speed of 100 Mbps and, as they use fiber optic cables, they provide connectivity over long distances. These networks have now been replaced by Fast Ethernet networks that provide the same speed and are more fault tolerant.