Token Ring

Local Area Network Access Methods

Token Ring

The second most popular type of LAN is the token ring. Almost 25% of all LANs worldwide are token ring LANS. Token Ring was originally developed by IBM, and have been since standardized by IEEE as IEEE 802.5. A token is a special control frame on token ring, token bus, and FDDI (Fiber Distributed Data Interface) networks that determines which stations can transmit data on a shared network. The node that has the token can transmit the data. Unlike contention-based networks, such as the Ethernet, workstations on token-based networks do not compete for access to the network. Only the station that obtains the token can transmit. Other stations wait for the token rather than trying to access the network on their own. On Ethernet networks, collisions occur when two or more workstations attempt to access the network at the same time. They must back off and try again later, which reduces performance, especially as the number of workstations attached to a network segment increases.

A token ring uses a ring topology, connecting all computers on the LAN in one closed loop circuit with each computer connected to the next. From the outside, a token ring appears to be a star, because all the cables flow into the central hub, but it is truly a ring, with messages passing from one computer to the next. All computers in a token ring LAN share the same common circuit, therefore, it is essential that access to the media be controlled by the data link layer. Token ring uses a controlled-access technique called token passing. The token a series of bits, travels between the computers in a predetermined sequence. A computer with a message waits to transmit waits until it receives a free token. The computer changes the free token to a busy token attaches its message to it, and retransmits it on the circuit to the next computer in the sequence. The computer receiving the message, changes the acknowledgment to ACK (or NAK) and sends the message back to the sender, who creates a new free token. One problem with token-passing protocols is dealing with lost tokens. If no token circulated through the network for a certain length of time or if a busy token circulates too often, the token monitor (designate one computer in the network) will create a new free token (and destroy the busy token if necessary).

Contention-Based

Contention-Based is the opposite of controlled access. Computers wait until the circuit is free (ex. no other computer is transmitting) and then transmit whenever they have data to send. Essentially, Contention-Based Protocols are "first come, first served." Contention is commonly used in Ethernet LANs and the most common example is Carrier Sense Multiple Access with Collision Detection (CSMA/CD).

Contention is method by which devices on the same shared multipoint circuit compete for time on the circuit. If no one is transmitting, a workstation can transmit. If someone else is transmitting, the workstation "backs off" and waits. Therefore, contention-based is a half duplex protocol. If two workstations transmit at the same time, a collision occurs. When the two workstations hear the collision, they stop transmitting immediately. Each workstation backs off a random amount of time and tries again. Workstations use a persistence algorithm to decide when to resubmit. The goal is for both workstations not to try again at the exact same time.

Almost 70% of all LANs in the world use Ethernet. The Ethernet LAN standard was originally developed by DEC, Xerox, and Intel, but has since become a formalized standard by the IEEE as IEEE 802.3. Ethernet uses a bus topology (a high speed circuit and a limited distance between the computers, such as within one building.) From the outside, an Ethernet LAN appears to be a star, because all cables connect to the central hub. Most Ethernet LANs span sufficient distance to require several hubs, but some LANs are build without the use of hubs (coax hubs). When several computers share the same communication circuit, it is important to control their access to the media. If two computers on the same circuit transmit at the same time, their transmission becomes garbled. The collisions must be prevented, or means to recover established. Like all contention-based techniques, CSMA/CD is very simple; wait until the bus is free and then transmit. A major advantage, a token doesn't need to be sent to transmit the data. Wireless LANs transmit data through the air (space) rather than through the wire or cable. Wireless LANs (IEEE802.11) CSMA/CA are similar to the Ethernet.

Overall: Good for bursty traffic, All stations contend for time, Distributed, Simple to implement, Efficient under moderate load, and tend to collapse under heavy load.

Network Architecture Qualities of Service

Connnectionless Message Routing vs. Connection-oriented Message Routing

Connectionless routing means each packet is treated separately and makes its own way through the network. It is possible that different packets will take different routes through the network depending on the type of routing used and the amount of traffic. Unlike connection-oriented routing, no connection is established. The sender simply sends the packets as separate, unrelated entities, and it is possible that different packets will take different routes through the network, depending on the type of routing used and the amount of traffic. Because packets following different routes may travel at different speeds, they may arrive out of sequence at their destination. The sender's network layer, therefore, puts a sequence number on each packet, in addition to information about the message stream to which the packet belongs. The network layer must reassemble them in the correct order before passing the message to the application layer. Connectionless is most commonly used when the application data or message can fit into one single packet.

Connection-oriented routing sets up a virtual circuit (one that appears to use point-to-point circuit switching) between the sender and receiver. The network layer makes one routing decision when the connection is established and all packets follow the same route. All packets in the same message arrive at the destination in the same order in which they were sent.

TCP/IP can operate either as connection-oriented or connectionless. When connection-oriented is desired, both TCP and IP are used. TCP establishes the virtual circuit with the destination and informs IP to route all messages