Computer Architecture

NETWORKS
The goal of any computer network is to allow multiple computers to communicate.
The type of communication can be as varied as the type of conversations you might have
throughout the course of a day
Networks Advantages and Disadvantages:

-Network Hardware, Software and Setup

Costs.
-Hardware and Software Management &
Administration Costs.
-Undesirable Sharing.
-Illegal or Undesirable Behavior.
-Data Security Concerns.

-Connectivity and Communication.

-Data SharingHardware Sharing.
-Internet Access.
-Data Security and Management.
-Performance Enhancement and Balancing.
-Entertainment.
NETWORK TYPES
Local area networks (LANs) were introduced to connect computers in a single office or
building.
Wide area networks (WANs) expanded the LANs to include networks outside the local
environment and also to distribute resources across great distances.
Personal area network (PAN) is a very small-scale network designed around one
person within a very limited boundary area.
Metropolitan area network (MAN), which is bigger than a LAN but not quite as big as
a WAN.
Network Interface Card (NIC)
The network interface card (NIC), also referred to as a network adapter card,
provides the physical interface between computer and cabling. It prepares data,
sends data, and controls the flow of data.
It can also receive and translate data into bytes for the CPU to understand.
NICs come in many shapes and sizes.
Different NICs are distinguished by the PC bus type and the network for which they
are used.
Let’s describe the role of NICs and how to evaluate them. (In following sections)
<<<<<<<<
 Compatibility of NIC
The first thing you need to determine is whether the NIC will fit the bus type of your
PC.

If you have more than one type of bus in your PC (for example, a combination PCI/PCI

Express), use a NIC that fits into the fastest type (the PCI Express, in this case).

This is especially important in servers because the NIC can quickly become a
bottleneck if this guideline isn’t followed.
Network Interface Card Performance

The most important goal of the NIC is to optimize network performance and
minimize the amount of time needed to transfer data packets across the network.

The key is to ensure that you get the fastest card that you can for the type of network
that you’re on.

For example, if your wireless network supports 802.11b/g/n, make sure to get an
802.11n card because it’s the fastest.
Physical And Logical LO3
Topologies
INTRODUCTION
Topologies
 Bus
 Ring
 Star
 Extended Star
 Mesh
 Hybrid
Physical Vs. Logical Topology

The actual layout of a network and its media is its Physical Topology

The way in which the data access the medium and transmits packets is the
Logical Topology.
Physical Vs. Logical Topology (2)

Your choice of Logical Topology will affect the Physical Topology –

and vice versa
Design carefully – it may be difficult to change part way through
the installation
Your choice will determine cable installation, network devices,
network connections, protocols

 CP2073 NETWORKING  11
FACTORS

Cost

Scalability

Bandwidth Capacity

Ease of Installation

Ease of fault finding and maintenance

BUS TOPOLOGY
BUS TOPOLOGY (2)

Network maintained by a single cable

Cable segment must end with a terminator

Uses thin coaxial cable (backbones will be thick coaxial cable)

Extra stations can be added in a daisy chain manner

BUS TOPOLOGY (3)

Standard is IEEE 802.3

Thin Ethernet (10Base2) has a maximum segment length of 200m

Max no. of connections is 30 devices

Four repeaters may be used to a total cable length of 1000m

Max no. of nodes is 150

BUS TOPOLOGY (4)
Advantages Disadvantages
Inexpensive to install No longer recommended

Easy to add stations Backbone breaks, whole network down

Limited no of devices can be attached
Use less cable than other
topologies Difficult to isolate problems

Works well for small networks Sharing same cable slows response rates
RING TOPOLOGY
RING TOPOLOGY (2)
No beginning or end (a ring in fact !!)
All devices of equality of access to media
Single ring – data travels in one direction only
Each device has to wait its turn to transmit
Most common type is Token Ring (IEEE 802.5)
RING TOPOLOGY (3)
Advantages Disadvantages
Data packets travel at great Requires more cable than a bus
speed A break in the ring will bring it
No collisions down
Easier to fault find
No terminators required
STAR TOPOLOGY
STAR TOPOLOGY (2)
• Centre point is a Hub
•Segments meet at the Hub
•Each device needs its own cable to the Hub
•Easy to maintain and expand
STAR TOPOLOGY (3)
Advantages Disadvantages
•Easy to add devices as the network expands •A star network requires more cable
than a ring or bus network
•One cable failure does not bring down the
entire network (resilience) •Failure of the central hub can bring
down the entire network
•Hub provides centralised management
•Costs are higher (installation and
•Easy to find device and cable problems
equipment) than for most bus
•Can be upgraded to faster speeds networks
•Lots of support as it is the most used
Extended Star Topology

A Star Network which

has been expanded to
include an additional
hub or hubs.
Mesh Topology (Web)
Mesh Topology (2)
oNot common on LANs
oMost often used in WANs to interconnect LANS
oEach node is connected to every other node
oAllows communication to continue in the event of a break in any
one connection
oIt is “Fault Tolerant”
Mesh Topology (3)
Advantages Disadvantages

Improves Fault Tolerance Expensive

Difficult to install

Difficult to manage

Difficult to troubleshoot
Hybrid Topology
HYBRID TOPOLOGY (2)

 Hybrid Topology – combines two or more different physical topologies

Commonly Star-Bus or Star-Ring

TYPES OF LOGICAL TOPOLOGY

Previous slides showed Physical Topologies

Only two Logical Topologies (Bus or Ring)

Logical Bus

•Modern Ethernet networks are Star Topologies (physically)

•The Hub is at the centre, and defines a Star Topology
•The Hub itself uses a Logical Bus Topology internally, to
transmit data to all segments
Logical Ring

Data in a Star Topology can transmit data in a Ring

Logical Ring (2)

CP2073 Networking 32
AN OVERVIEW OF
ADDRESSING ON NETWORKS
Four addressing methods:
 Application layer FQDNs, computer names, and host names
 Transport layer port numbers
 Network layer IP address
 IPv4 addresses have 32 bits and are written as four decimal numbers called octets
 IPv6 addresses have 128 bits and are written as eight blocks of hexadecimal numbers
 Data Link layer MAC address
 Also called physical address

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 34

MAC ADDRESSES
Traditional MAC addresses contain two parts
 First 24 bits are known as the OUI (Organizationally Unique Identifier) or block ID or company-ID
 Assigned by the IEEE
 Last 24 bits make up the extension identifier or device ID
 Manufacturer’s assign each NIC a unique device ID

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 35

MAC ADDRESSES

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 36

HOW IP ADDRESSES ARE
FORMATTED AND ASSIGNED
Two types of IP addresses:
 IPv4 - a 32-bit address
 IPv6 - a 128-bit address

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 37

HOW IPV4 ADDRESSES ARE
FORMATTED AND ASSIGNED
IPv4 addresses
 32-bit address organized into four groups of 8 bits each (known as octets)
 Each of the four octets can be any number from 0 to 255
 Some IP addresses are reserved

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 38

CLASSES OF IP ADDRESSES
IPv4 addresses are divided into five classes:
 Class A, Class B, Class C, Class D, and Class E

When class licenses were available from IANA:

 Class A license was for a single octet
 Class B license was for the first two octets
 Class C license was for the first three octets
 Class D and Class E addresses were not available for general use
 Class D begin with 224-239 and are used for multicasting and Class E begin with octets 240-254 and are used for research

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 39

CLASSES OF IP ADDRESSES

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 40

CLASSES OF IP ADDRESSES

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 41

HOW A DHCP SERVER ASSIGNS
IP ADDRESSES
Static IP addresses are assigned manually by the network administrator
Dynamic IP addresses are automatically assigned by a DHCP server
If a computer configured to use DHCP is unable to lease an IPv4 address from the
DHCP server
 It uses an Automatic Private IP Addressing (APIPA) address in the address range 169.254.0.1 through
169.254.255.254

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 42

PUBLIC AND PRIVATE IP
ADDRESSES
Class A, B, and C licensed IP addresses are available for use on the Internet
 Called public IP addresses

A company can use private IP addresses on its private networks

IEEE recommends the following IP addresses be used for private networks:
 10.0.0.0 through 10.255.255.255
 172.16.0.0 through 172.31.255.255
 192.168.0.0 through 192.168.255.255

NETWORK+ GUIDE TO NETWORKS, 8TH EDITION 43