This documents compares Ethernet against Token Ring networks. Both topologies have a lot to offer the market place, however, it is up to the organization to make an informed decision as to what is best for the company. Ethernet uses a bus topology in which all computers are connect to one straight line of cable, otherwise know as the backbone. Data is permitted to access the network at anytime, however, it can only enter when no other data is on the network. Data will travel down the backbone till it reaches its final destination or the end of the wire and eliminated. Token Ring works in a similar fashion except for the fact that the computers are connected in a loop. Data can only enter the network when given permission. At that time, data is transmitted around the ring until if reaches it final destination or eliminated by the monitoring node.
Ethernet vs. Token Ring
LANs are two or more computer that spans a small area such as a building or an office. The two types of Local Area Networks (LAN) we are going to discuss are Ethernet and Token Ring. Both of these networks are the most popular on the market today. Ethernet is the topology most often selected for network design but is it always the best design? Let’s take a look at both topologies and see what they have to offer us.
The Xerox Corporation developed the Ethernet in 1976. Ethernet uses the bus topology with data transfer rates of 10 Mbps. All nodes are connected to a single long cable. Any connected node can send and receive a signal along the cable. However before the node can send any data, it must listen to the network to see if it is in use. The source node will not release the data to the destination node until the network is clear. If two or more nodes send data at one time, a collision will transpire. Whenever this happens, the original data is dropped and resent by the source node.
Ethernet uses the carrier sense multiple access/collision detection (CSMA/CD) at the physical and link layer to detect collisions on the network. The frame structure for carrying data begins with alternating binary numbers that signals the destination node the frame is on its way. Following the preamble is the destination and source address for the data. The next section contains a two-byte field that is used to identify the data type. Then you have section that contains the actual data follow by the frame check sequence (FCS) field. The FCS contains cyclic redundancy check (CRC) value. The CRC is designed by the sending device and recalculated by the receiving node to check for damages that might have transpired during transit.
IBM created the Token Passing Ring Network and is the second most commonly implemented LAN, second only to Ethernet. Token Ring networks use token passing to send data across the network. The computers are grouped together and connected in a loop. The computer requesting to send data will need to obtain permission before transmitting any data. When the node obtains control of the network, like Ethernet, the data will be transmitted in a frame. The frame circles the network passing from one computer to the next until it reaches source of the transmission.
The frame format for the tokens noted above was obtained from www2.rad.com. It provides us with a breakdown of the token. The header, or start delimiter, alerts nodes of the tokens arrival. Next, the access control byte contains the priority and reservation field for access to the network. This is also where node would check to see if the data has been abandon on the ring for what could be an endless amount of time. The next section of the token is the control information. This area contains both the source and destination address. Finally, the last section of the token contains the End Delimiter, which signals the end of the frame.
Token Ring networks use a dedicated node and fault management mechanism for monitoring frames that are circling around the network without returning to its place of origin. This monitoring device will remove such frames and enhance performance of the network.
After all my research, reading and pondering, I have come to the conclusion that I have merely scratched the surface of the both Ethernet and Token Ring networks. Both topologies have a lot to offer an organization. The Ethernet bus is a simpler topology with less room to err. However, the Token Ring provides an organization with a network that has the ability to protect itself against failure. My personal preference is to employ a network that has less room to err, Ethernet, so that it is not concerned with protecting itself, Token Ring.