Sie sind auf Seite 1von 27

Lecture 7 Lecture 7

LAN Topologies LAN Topologies


Data Communications and Networks
LAN Topologies LAN Topologies
Data Communications and Networks
Semester 1
P d b Presented by:
Vajira De Silva B.Sc. (moratuwa), M.Sc. (melbourne)
LAN Transmission Categories LAN Transmission Categories
LAN transmissions fit into one of three categories:
Unicast
Multicast
Broadcast
24/12/2009
Data Communications & Networks - Lecture 7
2
Unicast Unicast
A single packet is sent from the source to a destination on a network. g
The source node addresses the packet by using the network address
of the destination node.
The packet is then forwarded to the destination network and the
network passes the packet to its final destination
24/12/2009
Data Communications & Networks - Lecture 7
3
Multicast Multicast
A single data packet is copied and forwarded to a specific subset of g
nodes on the network.
The source node addresses the packet by using a multicast address. p y g
24/12/2009
Data Communications & Networks - Lecture 7
4
Broadcast Broadcast
Frame is addressed to the broadcast address (a broadcast frame) and
is copied and forwarded across the network to all the devices.
Broadcasts are commonly found in LAN environments. y
Broadcasts do not traverse a WAN unless the Layer 3 edge-routing
device is configured to direct these broadcasts to a specified network
address.
An example of a data broadcast on a LAN could be a host searching
for server resources.
But, have to be cautious when used for multimedia broadcast traffic.
High bandwidth consumption
24/12/2009
Data Communications & Networks - Lecture 7
5
LAN Addressing LAN Addressing
LAN addresses identify individual or groups of devices. g
Addressing schemes vary depending on the protocol family and OSI
layer. y
MAC (Media Access Control) addresses
Network layer addresses
24/12/2009
Data Communications & Networks - Lecture 7
6
MAC Addresses MAC Addresses
MAC addresses identify network devices in LANs.
MAC addresses are unique for each LAN interface on a device.
48 bits in length and are expressed as 12 hexadecimal digits: g p g
First six hexadecimal digits, which are administered by the IEEE,
identify the manufacturer or vendor organizational unique identifier y g q
Last six hexadecimal digits comprise the interface serial number a
value administered by the specific vendor
MAC addresses are referred to as burned-in addresses (BIAs)
because they are burned into read-only memory (ROM).
MAC is defined at the data link layer (Layer 2) of the OSI model.
24/12/2009
Data Communications & Networks - Lecture 7
7
Network Layer Addresses Network Layer Addresses
Network layer addresses identify a device at the OSI network layer
(Layer 3).
Network layer addresses have two parts: the network of which the y p
device is a part and the device, or host, number of that device on
that network.
Devices on the same logical network must have addresses with the
same network part; however, they will have unique device parts.
The subnet mask determines where the network boundary ends and
the host boundary begins.
24/12/2009
Data Communications & Networks - Lecture 7
8
The Network Topology
Not in the lecture note !
The Network Topology
The network topology defines the p gy
layout of the network.
It shows how devices on the
network are interconnected.
Devices on the network are
termed nodes.
A network has both a physical and
a logical topology.
24/12/2009
Data Communications & Networks - Lecture 7
9
Physical vs Logical Topology
Not in the lecture note !
Physical vs. Logical Topology
Physical topology shows the physical plan of a network, which g
refers to the actual physical layout of the devices and media.
Logical topology refers to the paths that signals travel from one g p gy p g
point on the network to another.
24/12/2009
Data Communications & Networks - Lecture 7
10
LAN Topologies LAN Topologies
Four LAN topologies exist: g
Star (Hub-and-Spoke)
Ring
Bus
Tree
24/12/2009
Data Communications & Networks - Lecture 7
11
Star (Hub-and-Spoke) Topology (1) Star (Hub and Spoke) Topology (1)
All stations are attached by physical links to a central point, usually a
wiring hub or other device operating in a similar function.
Shielded twisted-pair (STP), unshielded twisted-pair (UTP), fiber- p p
optic cabling and wireless media can be used as physical links.
The star topology is the most commonly used architecture in p gy y
Ethernet LANs and resembles spokes in a bicycle wheel.
24/12/2009
Data Communications & Networks - Lecture 7
12
Star (Hub-and-Spoke) Topology (2) Star (Hub and Spoke) Topology (2)
24/12/2009
Data Communications & Networks - Lecture 7
13
Star (Hub-and-Spoke) Topology (3) Star (Hub and Spoke) Topology (3)
Advantages:
N bl t i i l i t f f il No cable segment is a single point of failure.
It is upgradeable, flexible, and reliable
It is easy to design and install It is easy to design and install
This topology makes diagnosing problems relatively easy since the
problem is localized to one computer or device
This topology allows for more throughput than any other topology
Disadvantages
Central hub is a single point-of-failure in that if it fails, every attached
station is out of service.
It requires a lot of cable to connect computers since a cable is required
between each device and the central location.
It is more expensive to build because of the additional cost of cables and devices It is more expensive to build because of the additional cost of cables and devices
like hubs and switches.
24/12/2009
Data Communications & Networks - Lecture 7
14
Ring Topology (1) Ring Topology (1)
All stations in a ring topology are considered repeaters and are
l d i l d i t enclosed in a loop no end points.
The repeater in this case is LAN-attached stations NIC.
B h NIC h LAN ll Because each NIC is a repeater, each LAN station will repeat any
signal that is on the network, regardless of whether it is destined for
th t ti l t ti that particular station.
Failure of a single NIC could bring the entire network down.
Some NIC controllers are capable of recognizing and handling the Some NIC controllers are capable of recognizing and handling the
defect and pulling itself off the ring, allowing the ring to stabilize and
continue operating. p g
Token Ring (IEEE 802.5) best represents a ring topology:
Physical topology (cabling) is considered to be a star topology. Physical topology (cabling) is considered to be a star topology.
But, it is a ring in logical topology.
24/12/2009
Data Communications & Networks - Lecture 7
15
Ring Topology (2) Ring Topology (2)
Although physical topology is a physical layer attribute, the media g g
access method used at the data link layer determines the logical
topology.
Token Ring (IEEE 802.5) defines a logical ring and contention!
A frame, called a token, travels around the ring and stops at each
d node.
If a node wants to transmit data, it adds that data and the addressing
i f ti t th f information to the frame.
The frame continues around the ring until it finds the destination
node which takes the data out of the frame node, which takes the data out of the frame.
The advantage of using this method is that there are no collisions of
data packets. p
24/12/2009
Data Communications & Networks - Lecture 7
16
Ring Topology (3) Ring Topology (3)
24/12/2009
Data Communications & Networks - Lecture 7
17
Ring Topology (4)
Not in the lecture note !
Ring Topology (4)
With single ring all the devices on the network share a single g g g
cable, and the data travels in one direction only.
With dual ring two rings allow data to be sent in both g g
directions.
This creates redundancy (fault tolerance) meaning that in the This creates redundancy (fault tolerance), meaning that in the
event of a failure of one ring, data will still be transmitted on
the other ring the other ring.
Fiber Data Distributed Interface (FDDI) uses light instead of
l t i it t t it d t d l i electricity to transmit data over a dual ring.
But, FDDI rings are physically cabled in a star topology
24/12/2009
Data Communications & Networks - Lecture 7
18
Ring Topology (5)
Not in the lecture note !
Ring Topology (5)
24/12/2009
Data Communications & Networks - Lecture 7
19
Bus Topology (1) Bus Topology (1)
Commonly referred to as a linear bus, all the devices on a bus y
topology are connected by one single shared cable segment.
Every station on this segment sees transmissions from every y g y
other station on the cable segment; this is known as a broadcast
medium.
The LAN attachment stations are definite endpoints to the
cable segment and are known as bus network termination points cable segment and are known as bus network termination points.
Ethernet (IEEE 802.3) best represents this topology. Ethernet has
the ability to utilize many different cable schemes the ability to utilize many different cable schemes.
24/12/2009
Data Communications & Networks - Lecture 7
20
Bus Topology (2)
Not in the lecture note !
Bus Topology (2)
24/12/2009
Data Communications & Networks - Lecture 7
21
Bus Topology (3) Bus Topology (3)
Advantages of a bus topology:
The thinnet cabling it uses is quite inexpensive.
It uses less cable compared to other physical topologies like star or
t d d t extended star
It works well for small networks
It does not need a central device, such as a hub, switch, or router It does not need a central device, such as a hub, switch, or router
Disadvantages of a bus topology:
It results in slower access to the network and less bandwidth due to the
sharing of the same cable by all devices
It is challenging to identify and isolate problems
A break at any point in the bus cable can disable the entire bus network
It needs terminators
24/12/2009
Data Communications & Networks - Lecture 7
22
Tree Topology (1) Tree Topology (1)
The tree topology is a logical extension of the bus topology and g g g
could be described as multiple interconnected bus networks.
The physical (cable) plant is known as a branching tree with all p y p g
stations attached to it.
The tree begins at the root and expands to the network endpoints. g p p
This topology allows a network to expand dynamically with only one
active data path between any two network endpoints.
A tree topology network does not employ loops in its topology.
Example: a bridged or switched network running the spanning tree p g g p g
algorithm, usually found with Ethernet (IEEE 802.3) networks.
24/12/2009
Data Communications & Networks - Lecture 7
23
Tree Topology (1) Tree Topology (1)
24/12/2009
Data Communications & Networks - Lecture 7
24
Not in the lecture note !
Other Types of Topologies
24/12/2009
Data Communications & Networks - Lecture 7
25
Mesh Topology Mesh Topology
The mesh topology connects all devices (nodes) to each other
for redundancy and fault tolerance.
It is used in wide-area networks (WANs) to interconnect LANs
and for critical networks.
The mesh topology is expensive and difficult to implement.
24/12/2009
Data Communications & Networks - Lecture 7
26
Hybrid Topology Hybrid Topology
The hybrid topology combines more than one type of topology. y p gy yp p gy
When a bus line joins two hubs of different topologies, the
configuration is called a star bus. g .
The bus line is used to transfer the data between the star
topologies topologies.
24/12/2009
Data Communications & Networks - Lecture 7
27

Das könnte Ihnen auch gefallen