The Ethernet standard defines a bus topology LAN.
|
In a bus topology, devices are directly connected to a central backbone.
|
Ethernet is a very common LAN technology.
|
The initial Ethernet specification was defined in 1980 by Xerox, Digital Equipment Corporation (Digital), and Intel.
|
The same group subsequently released the Ethernet 2 specification in 1984.
|
The IEEE 802.3 subcommittee adopted the Ethernet specification for its model.

As a result, Ethernet 2 and IEEE 802.3 are identical in the way they use the physical medium, though not compatible.
|
Ethernet operates at a baseband signaling rate of 10 Mbps.
|
There are three Ethernet wiring standards:

 10Base2
|
 10Base5
|
 10BaseT
|
The 10Base2 wiring standard, which is known as ^Rthin Ethernet^r, allows network segments up to 185 meters long using coaxial cable.
|
10Base5, known as ^Rthick Ethernet^r, allows network segments up to 500 meters long.
|
Both 10Base2 and 10Base5 provide access for several devices on the same segment.
|
Devices are attached by a cable that runs from an attachment unit interface (AUI) in the device to a transceiver that is directly attached to the Ethernet coaxial cable.
|
The 10BaseT wiring standard carries Ethernet frames on twisted pair wiring.
|
Twisted pair wiring is less expensive than coaxial cable.
|
Each device is individually attached to a hub using twisted pair wiring.
|
The hub in turn is attached to the coaxial backbone.
|
In this arrangement the hub is analogous to an Ethernet segment.
@@
An Ethernet LAN provides data transport across the physical link joining the devices on the LAN.
|
For example, an Apple Macintosh, an Intel-based PC, and a Cisco router can be directly attached over an Ethernet LAN.
|
Devices are identified on the Ethernet LAN by their MAC addresses.
|
The Cisco router uses the Cisco IOS interface type abbreviation ^Re^r followed by an interface number, for example 0 (zero), to indicate the 802.3 interface.
@@
When a device transmits data on an Ethernet segment, the transmission crosses the entire network.
|
Every device on the LAN receives the message.
|
Each device passes the message from its physical layer to its data-link layer.
|
At the data-link layer the destination of the message can be read.
|
If the message is not destined for a device, the device ignores it.
|
If the message is destined for a device, the device passes it up to its network layer.
|
When the message reaches the end of a network segment, it is absorbed by terminators to prevent it from reflecting back onto the segment.
|
If a device wants to send a broadcast - that is, a message to all devices on the LAN, it uses a data-link destination address comprising three octals of 1s (FFFF.FFFF.FFFF in hexadecimal).
|
Broadcasting on a LAN, when improperly used, can seriously impact on the performance of devices by interrupting them unnecessarily.
|
Therefore broadcasts should only be used when the MAC address of the destination is unknown or when the destination is all devices.
@@
The Ethernet specification describes a carrier sense multiple access/collision detection (CSMA/CD) LAN.
|
CSMA/CD works in the following way.
|
When a device wants to transmit, it first checks the network to see if any other device is currently transmitting.
|
If the network is not being used, the device proceeds with the transmission.
|
While the device is transmitting it continually monitors the network to ensure that no other device is transmitting.
|
Two devices may have data to transmit at the same time.
|
They both check the network and find that no one else is transmitting.
|
They both proceed to transmit.
|
When this happens, a collision occurs.
|
When a transmitting device recognizes a collision it transmits a jam signal that causes the collision to last long enough so that all other devices recognize it.
|
All devices that are currently trying to transmit will then stop sending frames for a random time before attempting to retransmit.
|
If subsequent attempts also result in collisions, the device will try to retransmit up to 15 times before giving up.
@@
After Ethernet, the next most popular LAN specification is Token Ring.
|
It was originally developed by IBM in the 1970s.
|
Token Ring specifications are administered by the IEEE 802.5 committee.

The 802.5 specification is almost identical to, and is completely compatible with, IBM's Token Ring.

Consequently, the term Token Ring is generally used to refer to both specifications.
|
Logically, a Token Ring network is a ring in which each station receives signals from its upstream neighbor and repeats them to its downstream neighbor.
|
Physically, however, Token Ring networks are laid out as stars, with each station connecting to a central hub called a multistation access unit (MAU or MSAU).
|
The stations are connected to the MAU using twisted pair wiring.
|
Typically each MAU connects up to eight Token Ring stations.
|
If a Token Ring network consists of more than eight stations, or if stations are located in different parts of the building, MAUs can be chained together to create an extended ring.
@@
Like Ethernet, Token Ring provides physical layer and MAC sublayer services.
|
However, it relies on the IEEE 802.2 LLC sublayer and upper layer protocols for point-to-point services.
|
Token Ring differs considerably from Ethernet in its use of the LAN medium.
|
Whereas Ethernet provides opportunistic access, Token Ring provides deterministic access.
|
No station can dominate the cable and therefore the network administrator can quite accurately determine and plan network performance.
|
^RExplorer Packets^r are used to locate a route to a destination through one or more source route bridges.
|
All stations on a Token Ring network are identified by their MAC address.
|
To indicate an 802.5 interface on a Cisco router you can use the Cisco IOS interface type abbreviation for token ring (^Rto^r) followed by an interface number (for example, 0).
@@
Token Ring networks operate by passing a small frame, called a ^Rtoken^r, around the network.
|
If a station has no data to send it simply passes the token on to the next station.
|
If a station has data to send it seizes the token by changing one bit in the token frame, the T bit.
|
It then inserts the data it wants to send onto the frame, and sends it to the next station in the ring.
|
The station can transmit one frame only when it receives a token.
|
The frame, with the data, circulates the ring until it reaches the destination station.
|
Here it is copied by the station, and the frame is tagged as having been copied.
|
The frame continues around the ring until it returns to the originating station.
|
Here, the data is removed from the frame and the token is released back to the network.
|
Originally only one frame could circulate on the ring at any one time.
|
^REarly token release^r allows multiple tokens to circulate on the ring.
|
In this case, when a station seizes the token and transmits its information frame it can release a new token onto the ring.
|
Due to the token passing and the fact that frames proceed serially around the ring, collisions are not expected on a Token Ring network.
@@
Some Token Ring networks use a priority system that permit certain stations to use the network more frequently.
|
This priority is controlled by two fields within the Access Control field of the token frame.
|
Three priority bits are used to designate the priority of the token.
|
Only stations that have equal or higher priority can seize the token.
|
Once seized and transmitted as an information frame, only stations with a higher priority than the transmitting station can reserve the token for the next pass around the network.
|
When the next token is generated, it is set to the priority of the reserving station.
|
Any station that raises a token's priority must reinstate it to the previous lower priority level after its transmission is complete.
@@
Just as in any network, a Token Ring network can suffer from errors.
|
Token Ring networks use a mechanism called an ^Ractive monitor^r to detect and compensate for such errors.
|
One station, which can be any station on the network, is selected to be the active monitor.
|
This station acts as a centralized source of timing information for other stations and performs several ring maintenance functions.
|
For example, if an originating station fails it cannot remove its frame from the network.
|
Without an active monitor station, this frame would continuously circulate.
|
This would prevent other stations from transmitting and would tie up the network.
|
The active monitor station can detect leftover frames and remove them from the ring.
|
Once it has done this, it generates a new token.
@@
When it sends information, an originating station sets two bits - the Address (or A) bit and the Copied (or C) bit - to 0 (zero).
|
When the frame returns it examines these two bits.
|
The A bit indicates whether or not the destination MAC address was recognized.
|
It is set to 1 by the destination station if it receives the frame.
|
The C bit indicates whether or not the data was copied.
|
It is set to 1 if the data has been copied.
|
Therefore, if the A and C bits are both set to 1, the destination station has copied the data.
@@
The Fiber Distributed Data Interface (FDDI) standards were published in 1987 by the American National Standards Institute (ANSI).
|
It was developed for high speed LANs with high bandwidth requirements such as video and CAD/CAM.
|
FDDI operates at 100 Mbps over optical fiber medium.
|
FDDI is therefore well suited to operating over large distances or where networks are located in electronically hostile environments, such as factory floors.
|
ANSI has developed a copper-based medium for 100 Mbps dual Token Ring LANs.

This new standard is known as Copper Distributed Data Interface (CDDI).
@@
The FDDI standard defines a Token Ring LAN operating on dual counter-rotating rings using a token passing protocol.
|
Data travels in one direction on one ring and in the opposite direction on the other.
|
Under normal operation the second ring acts as a backup only.
|
If a break occurs in the fiber on the primary ring, the break is bypassed and the data is passed onto the second ring.
|
The two rings can be used simultaneously to achieve a throughput of 200 Mbps.

This, however, will result in a lower fault tolerance.
|
Two types of devices are used on FDDI networks:

 dual attachment stations 
  (DASs), which are attached 
  to both rings
|
DASs are usually critical stations such as routers or mainframe hosts.
|
 single attachments stations 
  (SASs), which are connected 
  only to the primary ring by 
  way of a dual attachment 
  concentrator (DAC)
|
Critical stations can use a technique called ^Rdual homing^r to provide additional fault tolerance.
|
With dual homing, a station is single-attached to two DACs, providing an active primary link and a backup path to the FDDI LAN.
@@
FDDI is both logically and physically a ring topology.
|
It shares many features with Token Ring networks, for example, it uses a token passing protocol.
|
Just like in a Token Ring network, access to an FDDI network is determined by token possession.
|
However, unlike a Token Ring network, stations can attach a new token to the end of their transmission.
|
At any given time, several information frames can be circling the ring.
|
The FDDI frame format uses 4-bit symbols rather than 8-bit octets.
|
This means that the 48-bit MAC address for FDDI has twelve 4-bit symbols.
|
All FDDI stations, including routers, use MAC addresses.
|
To indicate an FDDI interface on a Cisco router you can use the Cisco IOS interface type abbreviation ^Rf^r, followed by an interface number, for example 0.
@@
FDDI networks are very reliable.
|
All stations monitor the ring for invalid situations such as a lost token, persistent data frames, or a break in the ring.
|
If a node does not receive a token from its nearest active upstream neighbor (NAUN) during a predetermined time, it transmits beacon frames to identify the failure and its domain.
|
If it receives its own beacon back, it assumes that the ring has been repaired.
|
If it does not receive the beacon back within a specified time, the DASs on each side of the failure domain loop, or wrap, the primary ring to the secondary ring to maintain network integrity.
@@
Finally, you'll explore ATM technology.
|
ATM (asynchronous transfer mode) is a high bandwidth, low delay technology for switching and multiplexing.
|
It is a cell relay technology capable of very high speeds.
|
It divides information to be transferred into fixed sized blocks called ^Rcells^r.
|
ATM is a scalable technology - it can adapt to networks of different sizes and topologies.
|
While ATM does not depend on any particular physical layer implementation, it requires a high-speed, high-bandwidth medium for optimal operation.
|
Optical fiber networks offer the perfect medium for ATM.
@@
ATM can support demanding applications such as

 interactive multimedia
|
 real-time video services
|
 client/server databases
|
 interconnection of existing 
  networks
|
ATM allocates bandwidth on demand, and so is particularly suited to applications whose transmissions are of a bursty nature.
|
ATM can be used in place of other network protocols (such as Ethernet and Token Ring).
|
Many of the services that ATM supports require much higher bandwidths than conventional network technologies can support.
@@
Like most new technologies, ATM has appeared in a variety of implementations.
|
The main impetus to the standardization of ATM has come from the International Telecommunication Union - Telecomms Standardization Sector (ITU-T) and the ATM forum.
@@
