Chapter 2-Network Basics

1
 Objectives of the Chapter
• Network topologies
• Structure of the Internet and its history
• Network media
• Protocols and organizations

2
 2.1 - Introduction
• Network – set of devices connected by
communication links.
• There are two possible types of connections:
point-to-point and multipoint
• Point-to-Point – provides a
dedicated link between two
devices. (E.g. Router-to
Router link in a WAN)
• Multiaccess (or Multipoint) –
more than two devices share a
single link. (E.g. Ethernet) 3
 2.2 – Network Topology

• Refers to the shape of a network, or the

network's layout
• There are four basic topologies:
– bus
– star
– ring
– mesh

4
 Bus Topology
• One long cable (the bus) links all nodes.
• Bus is terminated at both ends.
• Limited number of devices, distance between
nodes
• Advantages: Easy installation, cheap
• Disadvantages: Difficult reconfiguration, no fault
isolation, a fault or break in the bus stops all
transmission, signal reflection at the taps,
inefficient with heavy traffic

1-5
 Star Topology
• Central device, called hub or concentrator
• The hub:
– manages and controls all functions of the network.
– acts as a repeater for the data flow.

• Advantages:
– Less expensive than a mesh topology.
– Easy to install and reconfigure  additions, moves, and deletions involve only one
connection.
– It is robust.

• Disadvantages:
– A single point of failure, the hub.

– More cabling than in bus.

6
 Ring Topology
• All devices are connected to one another in the
shape of a closed loop
• Actually connected to a central device called MSAU
(multistation access unit), forming a star-wired ring
topology.
• Advantages:
– Relatively easy to install and reconfigure.
– Performance is even despite many users

• Disadvantages:
– Unidirectional traffic.
– Limited media length and number of devices.
– A break in the ring can disable the entire network.
7
 Mesh Topology
• Dedicated point-to-point link to every other node.
• Full mesh with n nodes has ½ n(n-1) cables; a node
needs to have (n-1) I/O ports.
• Advantages:
– Dedicated link  security & privacy
– Robust: Failure of link does not affect network
– Easy fault identification and isolation

• Disadvantages:
– Difficult installation and reconnection
– Requires large wiring space
– Expensive: I/O ports and cables

8
 Hybrid Topology
• A combination of the basic network
topologies.

9
 2.3 Network Categories
• Based on size, networks can be:
– LAN (Local Area Network).
• Small area, private media, high speed
• Equipment: hubs, switches, routers
• E.g. Ethernet, Token Ring, FDDI

– MAN (Metropolitan Area Network)

• Larger area than LAN, E.g. City
• May interconnect several LANs
• Main Equipment: Router
• E.g. Switched Multimegabit Data Service (SMDS)

– WAN (Wide Area Network)

• Covers large area (E.g. Different cities)
• Mostly media owned by third party, low speed
• Equipment: routers, Modems, WAN switches, etc.
• E.g. X.25, Frame relay, ATM, MPLS 10
Network Categories (Cont’d)

11
 The Internet
• internet – two or more networks connected
• Internet – global interconnection of
computers and networks
• 1957 - The US DoD formed an agency
called ARPA (Advanced Research
Projects Agency)
• 1969 - The US DoD commissioned the
fledgling ARPAnet for network research
• 1990 - ARPAnet ceased to exist and the
Internet effectively took its role
12
 The Internet (Cont’d)
• The Internet has no owner, but it follows accepted
standards and protocols.
 hardware and software manufacturers concentrate on product
improvements, knowing that the new products can integrate
with and enhance the existing infrastructure.
• Internet is organized into hierarchical structure of ISPs
• Tier 1 – national & international
connections. E.g. AT&T, NTT
• Tier 2 – often provide regional
service
• Tier 3 – locally provide services
to users

13
 2.4 Transmission Media
• Physical paths for the communication
signal.

14
 Guide media
• Coaxial Cable
• ThickNet (10Base-5)
– 50-ohm RG-8 and RG-11 cable
– vampire taps and transceiver cable

15
 10Base-5 Ethernet (Cont’d)
• 10Base-5 Ethernet is characterized by:
– 5/4/3 rule (Max. of 5 segments, connected with 4
repeaters, max. 3 segments populated)
– Max. segment length = 500 m
– Min. distance b/n tranceivers = 2.5 m
– Max. number of nodes / segment = 100
– Max. transceiver cable length = 50 m
– 50 ohm terminator at each end, one end
grounded
– Awkward to handle and install
16
 10Base-2 (ThinEthernet)
• ThinNet uses 50-ohm RG-58A/U coaxial cable
• BNC Connector

17
 10Base-2 Ethernet (Cont’d)
• 10Base-2 Ethernet is characterized by:
– 5/4/3 rule applies
– Max segment length = 185m
– Max devices per segment = 30
– Max.distance between connections = 0.5 m
– 50 ohm terminator at each end, one end
grounded
– Lighter and easier to install

18
• Advantages of coaxial cables:
– Highly insensitive to EMI
– Supports high bandwidths
– Heavier types of coax are sturdy and can
withstand harsh environments
– Mature technology

• Disadvantages of coaxial cables

– They are still vulnerable to EMI in harsh
conditions such as factories.
– Coax can be bulky
19
 Twisted Pair (TP) Cables
• Consist of color-coded pairs of insulated
copper wires ( = 0.4 to 0.8 mm) twisted
around one another.

• Twisting the wires reduces susceptibility to

crosstalk and noise.
• There are two types:
– Shielded Twisted Pair (STP), and
– Unshielded Twisted Pair (UTP) 20
• STP includes metal shielding over each
individual pair of copper wires.
• Shielding protects cable from external
Electromagnetic Interferences (EMI) 
can be used in noisy area.
• Standard for IBM Token Ring networks
and for Apple's LocalTalk.

21
• UTP:
– contains no shielding
– includes four pairs of wires enclosed in a common
sheath
– is a very flexible, low cost media, and can be used
for either voice or data communications
– is widely used for cabling LANs in todays comp.NK
– uses RJ connectors

22
• UTP cables are divided into different categories
(Cat 1-7):
– Each vary in the number of twists per foot and quality.
• Cat 3 cables have 2 twists per foot whereas Category 5 has
12.
• Higher categories allow transmitting data at higher speeds.
– Cat 1, with two pairs, is used for voice; cat 2-7 have 4
pairs
– Cat 2 is used for Local Talk at 4 Mbps Local Talk
– Cat 3 was used for 10 Mbps Ethernet and voice
– Cat 4 was briefly used for 16 Mbps Token Ring
– Cat 5 is used for 100 Mbps Ethernet (Fast Ethernet)
and 155 Mbps ATM
– Cat 6 is used for 1Gbps Ethernet
– Cat 7 is used for 10Gbps Ethernet
23
• Ethernet/IEEE 802.3 standards on UTP
include:

– 10Base-T - 10Mbps, 100m, RJ-45

– 100Base-TX (Fast Ethernet) - 100Mbps,
100m, RJ-45
– 1000Base-T (Gigabit Ethernet) - 1000Mbps,
100m, RJ-45
– 10GBase-T (10 Gigabit Ethernet) –
10000Mbps, 100m, RJ-45

24
• Advantages of TPs
– Telephone cable standards are mature
– Possible to use in-place telephone wiring if it
is of sufficiently high quality.
– UTP represents the lowest cost cabling.
• Disadvantages of TPs
– STP can be expensive and difficult to work
with.
– UTP may be unsuitable for use in high-EMI
environments.

25
 Fiber-Optic Cables (cont’d)
• Fiber-optic cables consist of:
– a fine cylinder of glass (core)
– surrounded by another glass cylinder (cladding). Cladding
has lower index of refraction than core.
– layers of protective materials
– transfers data in the form of light rather than electrical
signals.

26
• Two types of fiber-optic cables exist:
– SMF – single ray from a laser beam

– MMF – multiple rays from light-emitting

diodes (LEDs)

– carry numerous modes or light

rays simultaneously

27
Fiber-optic cables (cont’d)

28
• Single Mode Fiber (SMF)
– Small core, about 7-9 m in diameter
– Less dispersion
– Suited for long distance applications (up to
100 km)
• Multi Mode Fiber (MMF)
– Larger core than single-mode cable (50/62.5
m)
– Allows greater dispersion and therefore, loss
of signal
– Used for long distance application, but shorter
than single-mode (up to ~2km)
• For both MMF and SMF: Cladding is 125 m
29
• Advantages
– Lighter than copper cables
– Offer very high bandwidth
– Less attenuation  Longer cable run,
– Immune to EMI  suitable in noisy areas
– Greater immunity to tapping  highly secure
– Resistance to corrosive materials
• Disadvantages
– Require greater skill
– More expensive cable and interfaces

30
31
 Unguided Media
• Radiate electromagnetic energy from
antennae. This can be:
– directional (in a beam), or
– omni-directional, i.e. all around radiation
• Different Technologies:
– Infrared – for very short distance (e.g. remote
control)
– Radio wave – omni directional, 3kHz-1GHz
• Lower frequency signals are reflected off the
ionosphere
• above 30MHz are not reflected  line of sight
32
– Microwave
• Terrestrial
– line of sight of parabolic dish antenna at elevated site
– Long distance communication is passible using a Series of relay
stations d 7.14 K .h
– For voice and television transmission and private
communications and telephone networks, e.g. emergency
services, utilities etc.
– 2 to 40 GHz but is susceptible to attenuation (esp. the higher
range) and interference
– Attenuation can rise markedly in poor atmospheric conditions,
e.g. rain
• Satellite
– No line of sight problems
– can be used for point to point or broadcast transmission
– uses a downlink frequency of 4GHz and an uplink frequency of
6GHz
– Typical use: television distribution, long distance telephone
transmission, and private business networks for global
33
organizations
34
35
 2.5 Protocols and Standards
• A protocol is a set of rules that govern
data communications. It defines what is
communicated, how it is communicated,
and when it is communicated.
• The key elements of a protocol are syntax,
semantics, and timing
– Syntax – structure or format of data
– Semantics – meaning of each field
– Timing –when data should be sent and how
fast they can be sent.
36
 Protocols and Standards (Cont’d)
• Standards provide guidelines to
manufacturers, vendors, government
agencies, and other service providers to
ensure interconnectivity.
• Data communication standards fall into two
categories:
– De facto – are those that have not been approved by an
organized body but have been adopted as standards
through widespread use. --By fact or convention
– De jure – these have been legislated by an officially
recognized body. --By Law or Regulation
37
 Protocols and Standards (Cont’d)
• Data telecom. standards are set by:
– International Organization for Standardization (ISO)-standards
on a wide range of subjects
– International Telecommunication Union Telecommunication
Standards Sector (ITU-T)-standardize to achieve end-to-end
compatibility of international telecommunication connections
– American National Standards Institute (ANSI)-represents USA in
ISO.
– Institute of Electrical and Electronics Engineers (IEEE)-
standards for computing and communications.
– Electronic Industries Association (EIA)-define physical connection
interfaces and electronic signaling. Promotion of electronics
manufacturing concerns.

38
 Protocols and Standards (Cont’d)
• Internet Standards
• TCP/IP Protocol Suite standards are developed
and published by the Internet Architecture Board
(IAB)
• The IAB has two principal subsidiary task forces:
– Internet Engineering Task Force (IETF)
• responsible for publishing the RFCs (working notes of the
Internet research and development community). The final
decision of which RFCs become Internet standards is made by
the IAB, on the recommendation of the IETF.
– Internet Research Task Force (IRTF)
• promotes research of importance to the evolution of the future
Internet by creating focused, long-term and small Research
Groups working on topics related to Internet protocols,
applications, architecture and technology 39
Review Questions
1. Explain the pros and cons of the different physical topologies.
2. Of the different basic physical network topologies which one is most prevalent?
Which
ones does Ethernet commonly use?
3. Why is UTP more popular than the other copper wires?
4. In what ways is optical fiber superior to the other media types?
5. Explain the 5/4/3 rule.
6. Ethernet has the following specifications:
a. 10Base5
b. 10Base2
c. 10BaseT
d. 100BaseTX
Explain what each part of the above specifications signifies.
7. What problem is solved by using the graded index multimode fiber instead of the step
index multimode fiber? Explain how the problem is solved.
8. What is/are the difference(s) between microwave and radio transmission?
9. Describe the three key elements of protocol.
10. Why do we need standards? What is the difference between de facto standards and
de jure
standards? Who is responsible for setting Internet standards?
11. Assume six devices are arranged in a mesh topology. How many cables are needed?
How many ports are needed for each device? 40
Objectives
• Circuit switching and packet switching
• Network layered architecture
• delay and loss

41
2.6 – Network Switching and Multiplexing
• Transmission over short distance:
– Point-to-point (direct connection) or LAN
• Transmission over long distance:
– switching

42
• Circuit Switching:
– A physical path
established between
end systems.
• E.g. Telephone system
– Three phases:
• Setup Phase – circuit established
• Data Transfer Phase
• Teardown Phase – release resources
– Advantages:
• Fixed data rate and hence fixed delays
• Guaranteed continuous delivery
– Disadvantages:
• Inefficient for bursty data
• Difficult to support variable data rates
• Inefficient for short messages (call setup)
43
• Message Switching:
– No physical path is established
– when the sender has a block of data to be
sent, it is stored in the first switching office
(i.e., router) and then forwarded later, one
hop at a time.
– Each block is received in its entirety,
inspected for errors, and then retransmitted.
– Messages can be very large, this can cause
buffering problems and high delay times
44
• Packet Switching:
– Messages broken into discrete units called
packets
– Each packet contains data and headers
– Headers can contain information about
source and destination address
– Store and forward, on a packet basis
– Two approaches:
• Datagram– mainly used in the Network Layer
• Virtual Circuit– a Data Link Layer technology

45
– Datagram packet switching
• Provides connectionless service
– no connection established before transmission of packet
• Each packet may be individually routed
• May take different routes and arrive out of order
• Reassembled at the destination
• Advantages:
– Efficient for bursty data
– Easy to provide bandwidth on demand with variable
rates
• Disadvantages:
– Variable delays
– Difficult to provide QoS assurances (Best-effort service)
– Packets can arrive out-of-order

46
• Virtual Circuit packet switching
– It is connection-oriented, which merges datagram packet
switching and circuit switching to extract advantages of both
– Packets flow on logical circuits; no physical resources are
allocated
– Each packet contains a virtual circuit identifier (VCI) instead
of destination address
– A circuit identifier is local to a link and is updated by each
switch on the path of the packet from its source to its
destination
– No routing decisions required for each packet
– Not a dedicated path
– A virtual circuit is defined by the sequence of the mappings
between a link taken by packets and the circuit identifier
packets carry on this link. This sequence is set up at
47
connection establishment time.
Virtual Circuit packet switching

Operating Principle of the VC 48

 Virtual Circuit packet switching
• Advantages:
• Packets are delivered in order, they take the same route;
• Overhead to packets is smaller, do not contain full
address;
• Connection is more reliable, network resources are
allocated at call setup;
• Billing is easier, billing is per call and not per packet.
• Disadvantages:
• The switching equipment needs to be more powerful,
since each switch needs to store details of call;
• Resilience to the loss of a trunk is difficult, if a failure,
calls are dynamically reestablished over a different route.

• Examples: X.25 and Frame Relay. 49

 Virtual Circuit Vs Datagram
• Virtual circuits
– Provide sequencing (in order delivery)
and error control
– Packets are forwarded more quickly
o No routing decisions to make
– Less reliable
o Loss of a node looses all circuits through that
node
• Datagram
– No call setup phase
o Better if few packets
– More flexible
o Routing can be used to avoid congested
parts of the network
50
 Sharing a Media: Multiplexing
• One communication line may be used to
transmit different data using multiplexing
techniques.
• Circuit switching uses FDM or TDM.
• In FDM, the frequency spectrum is divided into
frequency bands, with each user having
exclusive possession of some band.
• In TDM, the users take turns (in a round-robin
fashion), each one periodically getting the entire
bandwidth for a little burst of time.
51
 Sharing a Media: Multiplexing
• Packet switching uses statistical multiplexing.
• In statistical time division multiplexing, time is
divided among different sources based on demand and
packets from different sources are
interleaved on link.
• Packets contending for the link are buffered. Buffer
overflow is called congestion.
• Sequence of A & B packets does not have fixed pattern,
bandwidth shared on demand ⇒
statistical multiplexing.

52
Sharing a Media: Multiplexing

53
Packet switching versus circuit switching

• Packet switching is not suitable for real-time

services like telephone calls and video
conferencing because of its variable and
unpredictable end-to-end delays.
• On the other hand, packet switching:
1. offers better sharing of bandwidth than circuit
switching, and
2. is simpler, more efficient, and less costly to
implement than circuit switching
3. is great for bursty data.

54
 Multiplexing
• Examples
1) Suppose users share 1 Mbps link, each user
sends data at 100kbps when “active”, and active
for 10% of time. Hence, only 10 users can be
“active” at the same time.
• Circuit switching with TDM:
– Only 10 users can connect at a time
– Every time a user is not active, the time slot is wasted!
• Packet switching with statistical multiplexing:
– With 35 users, probability that more than 10 users are
active at the same time is less than .0004; thus 35 users
can connect at a time

55
 2) Consider the figure below, where L = 7.5 Mbits and
R = 1.5 Mbps
– For message switching:
• Delay = 3L/R =3(7.5)/1.5 = 15 sec

– For packet switching with 5000 packets:

Size of each packet 7.5 x10 6 bits 5000 1500 bits

Time to transmit one packet on one link L / R 1500 bits 1500 mbps 10  3 s 1ms
To transmit all packets through the first link 5000 x10  3 s 5s
Time to transmit the last packet through the last 2 links 2ms 0.002 s
Total time required to transmit the message 5.002 s 56
 2.7 – Network Performance

• Network performance
– How good is the network?
• Performance metrics include:
– Bandwidth, Throughput, Link utilization
– End-to-end delay
– Packet loss
– Delay-variation(or Jitter)
– Bandwidth-Delay Product

57
 Performance Metrics
• Bandwidth – two contexts:
– Bandwidth in hertz
• the range of frequencies in a composite signal or the
range of frequencies that a channel can pass.
– Bandwidth in bits per second (or Capacity)
• maximum rate at which data can be transmitted
through a channel or link.
– Throughput is the actual rate at which data is
transmitted.
– Link utilization = throughput of the link / link rate

58
 End-to-End Delay
• Packet Delay or Latency -the difference in time between when the
signal is transmitted, and when it is received.
• Delay consists 4 components:
– Processing Delay (Tproc) – time for checking errors and
determining next link and direct to a queue.
– Queuing Delay (Tque) – time between when the processing delay is
over and start of transmission
– Transmission delay (Ttrans) - time required to push all of the
packet's bits into the wire (L/R)
– Propagation delay (Tprop) - time a signal takes to reach destination
node once transmitted by transmitter (d/s) where s= speed of
propagation (m/s)
– Thus, total delay (Ttot) is given by:
• Ttot= Tproc + Tque + Ttrans + Tprop

59
 Switching Delay

• Circuit switching • Packet switching

– Connection – Store-and-
establishment forward delay
delay – Propagation
– Propagation delay delay
– Processing delay – Processing delay
(typically small) – Transmission
– Transmission delay
delay – Queuing delay
60
 Transmission Vs Propagation Delay
• Group of cars analogy
– Given: a group of 10 cars, moving at 100 km/hr.
A petrol station takes 12 sec to service a car
(transmission time).
• Analogy: car ~ bit; group ~ packet.
– How long will it take until the group of cars is lined
up before the 2nd petrol station?
– Solution
• Time to “push” the entire group from 1st petrol station
onto highway = 12*10 = 120 sec
• Time for last car to propagate from 1 st to 2nd petrol
station: 100km/(100km/hr)= 1 hr
• Total time = 62 minutes (1 hr + 120sec)

61
 Transmission vs Propagation Delay
(cont’d)
• Suppose the cars “propagate” at the rate of 1000
km/hr and petrol station takes 1 min to service a
car. Will cars arrive at 2nd station before all cars
are serviced at 1st station?

Solution
• After 7 min (1 min  1000
100km
km / h
*
60 min
1hr
), 1 st
car will be at
2nd station and 3 cars are still at 1 st station.

62
 Delay Analysis
Circuit Switching:
Assume:
Number of hops = M
Per-hop processing delay = P
Link propagation delay = L
Transmission speed = R bit/s
Message size = S bits
Total Delay = total propagation +
total transmission +
total processing
= 4ML + S/R + (M-1)P

Total Delay excluding time for

acknowledgment:
= 3ML + S/R + (M-1)P

63
Datagram Packet Switching:
Assume:
Number of hops = M
Per-hop processing delay = P
Link propagation delay = L
Packet transmission delay = T
Message size = N packets
Total Delay = total propagation
+ total transmission
+ total store & forward
+ total processing
= ML + NT + (M-1)T + (M-1)P

64
Virtual Circuit Packet Switching:
Assume:
Number of hops = M
Per-hop processing delay = P
Link propagation delay = L
Packet transmission delay = T
Message size = N packets
Total Delay = total propagation
+ total transmission
+ total store & forward
+ total processing
= 4ML + NT + (M-1)T + 4(M-1)P
Excluding time for acknowledgement:
= 3ML + NT + (M-1)T + 3(M-1)P

65
Q1. Almaz and Belay are 4 hops apart on a datagram packet-switched network
where each link is 160 km long. Per-hop processing delay is 5s. Packets
are 1500 bytes long. All links have a transmission speed of 56kbps (kbit/s).
The speed of light in the wire is approximately 200,000 km/s. If Belay sends
a 10-packet message to Almaz, how long will it take Almaz to receive the
message up to the last bit (measured from the time Belay starts sending)?

Solution: We know the following:

– Number of hops M=4,
– Number of packets N=10,
– Per-hop processing delay P=5s=0.000005s,
– Link propagation delay L = distance/speed of light = 160/200,000 = 0.0008s,
– Packet size = 1500 bytes = 1500*8=12,000 bits,
– Packet transmission delay T = packet size/transmission speed = 12,000/56000
=0.214s.
Delay = ML + NT + (M-1)T + (M-1)P =0.0032 + 2.14 + 0.642 + 0.000015 =
2.785s.

Note that the total delay is dominated by the transmission delay which
depends on link speed. A link with a higher transmission speed can reduce
the delay dramatically.

66
Q2. Now, suppose the link transmission speed is 1Gbps (Gbit/s). How
long will it take Almaz to receive the message up to the last bit
(measured from the time Belay starts sending)?

Answer: As before, we know the following:

– Number of hops M=4,
– Number of packets N=10,
– Per-hop processing delay P=5ms,
– Link propagation delay L = distance/speed of light = 160/200,000 =
800s,
– Packet size = 1500 bytes = 1500*8=12,000 bits,
– Packet transmission delay T = packet size/transmission speed =
12,000/109 =12s.
Delay = ML + NT + (M-1)T + (M-1)P =3200 + 120 + 36 + 15 =
3371s = 3.371ms.

Note that the total delay is now dominated by the propagation delay
which cannot be improved because it is constrained by the speed of
light. Hence, it is unlikely that future technologies will significantly
reduce the delay of Belay’s message at this point.

67
Q3. Repeat Q1 and Q2, assuming that the network uses circuit switching
instead of datagram packet switching. Belay’s message is the same length
as before.

Answer: Year 1989:

– Number of hops M=4,
– Message size S = 10 * 1500 * 8 =120,000 bits (it is not packetized)
– Link transmission speed R = 56kbit/s,
– Per-hop processing delay P=0.000005s,
– Link propagation delay L = distance/speed of light = 160/200,000 = 0.0008s,

Delay = 3ML + S/R + (M-1)P =0.0096 + 2.14 + 0.000015 = 2.1496s

Note that the delay improved over the case of datagram packet switching for
the same link speed. Why? (No store and forward delay)

Now, let link transmission speed be R= 1Gbp

Delay = 3ML + S/R + (M-1)P =9600 + 120 + 15 = 9735ms = 9.735ms

Note that the delay is worse than in the case of datagram packet switching.
Why? (Because of the propagation delay)

68
 Packet Loss
• Failure of one or more packets to arrive at
their destination
• Effects of packet loss:
– Mutilation of received data (errors)
– Broken-up images
– Unintelligible speech
– The complete absence of a received signal
• Cause of packet loss:
– Excessive attenuation and system noise
– Hardware failure and software corruption
– Buffer overflow (congestion)
69
 Delay Variation
• Delay variation (or jitter) refers to the
variation in the packet arrival time; it should
be kept minimum.

• Audio and video are very sensitive to jitter;

data is tolerant

70
 Bandwidth-Delay Product
• Suppose the bandwidth of a link is B bps.
• Also, suppose a bit needs Td to move from
the source to the destination
• During time Td,
B.Td bits will be in the link
• Thus, the bandwidth-delay product defines
the number of bits that can fill the link.

71
 2.8 Network Architecture
• A Network Architecture is a structured set of
protocols that implement the exchange of
information between computers
• The architecture is layered to reduce
complexity.
– Each layer offers certain services to the layer
immediately above it.
– Each layer shields the higher layer from the
details of implementation of how the services
are offered.
– Layer "n" on one station carries on a
conversation with layer "n" on another
network station.
72
 Network Models (Cont’d)
• The messages exchanged between
network entities are called protocol
data units (PDUs)
• PDU at layer n is called n-PDU
• Layer-n protocol defines contents and
format of an n-PDU as well as the
manner in which the n-PDUs are
exchanged
• Protocols of the various layers, taken
together, are called protocol stack
73
 Network Models (Cont’d)
• In a computer network, each layer may
perform one or more of the following tasks:
– Error control
– Flow control
– Segmentation and reassembly
– Multiplexing
– Connection setup
– Routing

74
 The OSI Model
• Two Network Models: OSI and TCP/IP
• OSI (Open System Interconnect) has 7
layers:
– Layer 7: Application Layer
– Layer 6: Presentation Layer
– Layer 5: Session Layer
– Layer 4: Transport Layer
– Layer 3: Network Layer
– Layer 2: Data Link Layer
– Layer 1: Physical Layer
75
The OSI Model (Encapsulation)
Application
Application Data Data Application
Header of
Application
Application Application
Layer
Layer AH Data Layer

Presentation Presentation
Layer
PH AH Data Layer

Session Session
Layer SH PH AH Data Layer

Transport Transport
Layer TH SH PH AH Data Layer

Network NH TH SH PH AH Data Network

Layer Layer

Data Link Data Link

Layer DH NH TH SH PH AH Data Layer

Physical Physical
Layer Bits Layer

76
 OSI Model Layers
• Layer 7: Application
• Provides network access to application programs
• Example: File Transfer, Electronic Mail
• Layer 6: Presentation
• Conversion of data format (e.g. text to pop-up
window)
• Encryption/decryption (e.g. Secure sockets)
• Layer 5: Session
• Support the dialog between cooperating application
programs
• Session management, synchronization
• Example: RPC
77
 OSI Model Layers (Cont’d)
• Layer 4: Transport
– Controls delivery of data between hosts
– Connection management, error control, flow control, multiplexing
– E.g. TCP, UDP.
• Layer 3: Network
– Moves packets inside the network
– Performs routing, addressing, switching
– Example: IP
• Layer 2: Data Link
– Reliably transfers frames over a link
– Performs synchronization, error control, flow control
– Example: PPP
• Layer 1: Physical
– Converts bits into electrical or optical signals
– Transmits these signals over the hardware communication medium
– Example: RS-232

78
 The TCP/IP Protocol Suite
• First defined in 1974
• The protocol architecture
of the Internet
• The TCP/IP suite has four
Application
layers: Layer telnet, ftp, email

– Application Transport
TCP, UDP
– Transport Layer

– Internet Internet IP, ICMP, IGMP

– Network Interface (Access)
Network
Device Drivers
Interface

79
 TCP/IP Protocol Layers (Cont’d)
Layer # Stack PDU Purpose

Layer 5 Application Message Supports network applications. It includes

many protocols: HTTP, SMTP and FTP to
support file transfer.
Layer 4 Transport Segment Responsible for transporting application-
layer messages between the client and
server sides of an application. In the
Internet, there are two transport protocols:
TCP and UDP.
Layer 3 Network Datagram Responsible for routing datagrams from one
host to another. Two principal components:
IP protocol and routing protocols.
Layer 2 Network Frame Responsible for repackaging the data
Interface into a frame for final physical
80
transmission. E.g. Ethernet and PPP.
 TCP/IP Encapsulation
User data
Application
Application
Header User data
TCP
TCP Header Application data
IP TCP segment

IP Header TCP Header Application data

Ethernet
IP datagram
Driver
Ethernet Ethernet
Header
IP Header TCP Header Application data Trailer

Ethernet frame

81
OSI vs TCP/IP models

82