Download What is data communication? - Information Systems

yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Document related concepts

Low-voltage differential signaling wikipedia, lookup

Asynchronous Transfer Mode wikipedia, lookup

Distributed firewall wikipedia, lookup

Piggybacking (Internet access) wikipedia, lookup

CAN bus wikipedia, lookup

Wake-on-LAN wikipedia, lookup

List of wireless community networks by region wikipedia, lookup

Network tap wikipedia, lookup

Computer network wikipedia, lookup

Deep packet inspection wikipedia, lookup

Zero-configuration networking wikipedia, lookup

Cracking of wireless networks wikipedia, lookup

Internet protocol suite wikipedia, lookup

Airborne Networking wikipedia, lookup

Recursive InterNetwork Architecture (RINA) wikipedia, lookup

IEEE 1355 wikipedia, lookup

UniPro protocol stack wikipedia, lookup

Data Communication
What is data communication?
• Data communications deals with the
transmission of signals in a reliable and efficient
• Ultimately, it’s about transmitting data (i.e., bits)
across some physical transmission medium:
– Electricity - copper wire, twisted pair, undersea cable
– Light - infra-red through air, laser through fibre-optic
– Electromagnetic radiation - radio, microwave, satellite
What is understood by the term
The term communication is defined as the act
of disseminating information.
It presupposes that:
there is information to disseminate
the desire or requirement to disseminate exists
there is an agency to send/transmit information
there is a means of encoding information
there is a medium to carry the information
there is a recipient to receive the information
the recipient is capable of understanding the
information received
• Let us generalize
• In a face-to-face
the process just
conversation between
two individuals
• In any
following takes place:
1. Conversion of brain waves into speech.
2. Agreement of both individuals on which
vocabulary to use.
3. Agreement of both individuals on volume
level at which both can be heard
4. Agreement of both individuals on the
rate of talking at which each can
understand the other’s speech.
5. Agreement of both individuals on the
rules used to decide when to speak and
when to listen, i.e. how the flow of
information is managed.
6. Conversion of the audio signals into
brain waves.
between two
entities the
following properties
are required:
Signal compatibility
Signal strength
Data rate
Communication Model
• Source
– generates data to be transmitted
• Transmitter
– Converts data into transmittable signals
• Transmission System
– Carries data
• Receiver
– Converts received signal into data
• Destination
– Takes incoming data
Diagram of Simplified Communication
Key Communications Tasks
Transmission System Utilization
Signal Generation
Exchange Management
Error detection and correction
Addressing and routing
Message formatting
Network Management
Simplified Data Communications Model
Basic Elements of a Communication
• In any communication between two entities the
following 10 elements can be identified:
A Sender.
A Receiver.
Addressing, to identify where the Receiver is.
Protocol – a set of co-operation rules to achieve
5. Transmission code - an agreed “language” to be used.
6. Transmission rate - the speed at which “what is being
communicated” is being sent.
7. Transmission synchronisation - how to recognise what is
being communicated.
8. Transmission medium.
9. Error detection and correction.
10.Transmission efficiency - how much overhead must be
added to manage the transmission.
Transmission Media
• Two wire -“telegraph wires” seen in old films.
• Simplest arrangement, with two wires, separated by air.
• Can pick up interference, and suffer “crosstalk”.
• Only reliable for low data rates.
• Twisted Pair - currently used for domestic
• Two insulated wires twisted together.
– Any interference affects both wires equally.
– May also have an additional protective screen of metallic foil –
“shielded twisted pair”.
– Suitable for short distance medium speed links.
– Suffers from “skin effect”, leading to higher resistance at higher
data rates.
“Skin effect” – HF signals carried only on skin of wire, in effect reducing the
area of the wire from a solid wire to a tube of the same diameter.
Transmission Media
• Coaxial cable - commonly seen on TV aerial leads
– Single central wire, separated from woven outer
conductor by plastic insulation.
– Not prone to interference.
– Can support medium to high data rates.
• Optical Fibre
– Similar to coaxial cable in appearance:
» Uses single strand of glass as core, with light shield around it.
– Immune to electrical interference , and difficult to
» Often used in industrial or other electrically “noisy” environments.
– Capable of high data rates
– Mechanically weaker than electrical wires, and difficult to
Transmission Media
• Microwaves -ultra high frequency radio waves
– Line of sight from sender to receiver.
– No need for wires, so good across rivers, or main roads
– Extremely high data rates
– Satellite microwaves:
– Mainly through space so long lines of sight.
– Little human interference, but affected by extreme solar
– Terrestrial microwaves:
– Need repeater stations if lines of sight short
» Curvature of earth, or mountains, or buildings
Data Transmission Terminology
• Transmission may be simplex, half-duplex or
– Simplex – in one direction only.
– Half-duplex – in both directions, but only in one direction
at any time.
– Full-duplex – in both directions simultaneously, if
• Transmission media may be guided or unguided.
– Guided – the medium is bounded and the transmission
contained within it (e.g. fibre-optic or electrical cable)
– Unguided – the medium is unbounded (e.g. radio waves
in the air, or in space).
Data Transmission Terminology
• In a direct link, (or data link), a transmission
– Propagates signals directly from transmitter (sender) to
– With no intermediate devices.
» except amplifiers (or repeaters) to increase signal
• In guided transmission media
– A configuration is point-to-point if it provides a direct link
between two devices, and those are the only two devices
sharing the medium.
– A configuration is multipoint, if more than two devices
share the same medium.
Data Transmission Terminology
Transmitter /
Receiver A
Transmitter /
Receiver B
(a) Point-to-point
Transmitter /
Receiver A
Transmitter /
Receiver B
Transmitter /
Receiver C
Transmitter /
Receiver D
Connection with amplifier
(b) Multipoint
Guided transmission configurations
Data Encoding
• Encoding means changing how data are
– This can be for convenience:
– Morse code alphabet used in early radio transmissions.
– Encoding to hide the meaning of data is “encryption”.
• Computer data are represented in an encoded
form for storage or transmission within and
between computers.
– The most common codes used to store digital data are:
» ASCII (American Standards Committee for Information
» EBCDIC (Extended Binary Coded Decimal Interchange
Data Encoding
• Data are transmitted using electromagnetic
• Data exists in analogue or digital forms.
• Analogue or digital data can be encoded using
either analogue or digital signals.
– For example digital data can be transmitted using analogue
– The telephone network traditionally used analogue signals to
represent voices.
– The telephone network was well-established when transmission
of digital computer data became necessary.
• The latter allows normal computer
communications using widely available
telephone lines.
– This is achieved using Modems.
Analog vs. Digital Signal
Data Encoding
Transmission direction
analogue signals
digital signals
Modulation is the conversion of the digital signal into an analogue signal, and
demodulation converts the analogue signal back into a digital signal. These
processes are carried out by a Modem.
Another device used is a Codec (coder-decoder).
All transmissions occur within a range of frequencies called the Bandwidth.
Signalling Technologies
• Baseband is the transmission of digital signals
without modulation.
• In a baseband communication network, digital signals (0s
and 1s) are put onto the medium as voltage pulses.
• The entire bandwidth is consumed by the signal.
• Broadband uses coaxial cable to provide data
transfer by means of analogue signals.
• The bandwidth is divided in different frequency bands or
• In a broadband communication network involving computers,
digital signals are passed onto the medium through a modem
and transmitted over one of the channels. So, several
different communication networks can be implemented over
the same medium.
Signalling Technologies
• Analogue transmission is used to mean the
transmission of analogue signals without regard
to their content.
• Digital transmission, on the other hand, is used
to mean the content of the signal.
Data Transmission : Data and Signals
Analogue Data
Digital Data
Analogue Signal
Digital Signal
Two alternatives:
Analogue data are encoded
1) Signal occupies the same using a codec to produce a
spectrum as the analogue data;
digital bit stream
2) Analogue data are encoded to
occupy a different portion of
Digital data are encoded using a
modem to produce analogue
Two alternatives:
1) Signal consists of two
voltage levels to represent
the two binary values
2) Digital data are encoded
to produce a digital signal
with desired properties
Data vs. Signal
Data Transmission: Treatment of Signals
Analogue Signal
Digital Signal
Analogue Transmission
Digital Transmission
Is propagated through
amplifiers. Same treatment
whether signal carries
analogue or digital data
Assumes that the
analogue signal carries
digital data. Signal is
propagated through
Not used
Digital signals represent
a stream of 0s and 1s.
Signal is propagated
through repeaters
Transmission Synchronisation
• Synchronisation is essential for transmitter and
receiver to understand each other.
• In serial transmission the following types of
synchronisation are required:
1. Bit synchronisation - how to detect each bit.
2. Byte or character synchronisation - how to group the
bits to make a character or byte.
3. Block synchronisation - how to group the
characters/bytes to make a block (a frame or a packet)
• Bit synchronisation depends on how the signal is
Transmission Synchronisation
• In serial transmission there are two standard
ways of achieving character and block
• Asynchronous Transmission or Character
– The time interval between characters is random.
– Each character is synchronised by the use of a start
bit, and either one or two stop bits.
– The bit rate is constant on a per character basis
Transmission Synchronisation
• Synchronous Transmission or Block
– Each block is synchronised by the use of a number of
synchronisation characters that are transmitted first
– These are followed by a start of block character,
which is followed by the data block, and transmission
is finished with an end of block character.
– The bit rate is constant for the whole transmission of
the block I.e. the time interval between characters is
Asynchronous vs. Synchronous
Asynchronous Transmission
Synchronous Transmission
Start and Stop bits reduce efficiency
More efficient use of bandwidth
A character can be transmitted at Characters buffered into blocks for
random times
Variable idle time between characters No idle time between characters
Constant bit rate within a character.
No limit on block length
Low speed communications (19.2
Synchronisation errors result in loss
of only a single character
Constant bit rate over a block. Blocks
limited to a maximum size
Higher speed communication
( 10 Mbits/s)
Synchronisation errors result in loss
of a complete block
Data Transmission: Modes
• Computer based communications always use
Digital Transmission,
– What is transmitted is digital data, using either an
analogue or digital signal.
– Normally, the digital data are recovered and repeated
at intermediate points to reduce the effects of noise.
• Irrespective of the type of communications
facility being used, in most applications data are
transmitted between computers in a bit-serial
mode,more commonly known as serial
Data Transmission: Modes
• Within a computer, data are transferred in a
word-parallel mode, most commonly known as
parallel transmission.
• In computer communications is necessary to
perform a parallel-to-serial conversion, in the
transmitter, serial-to-parallel conversion in the
• These conversions are done in the computer
interface to the network
Transmission efficiency
• Extra bits (start and stop bits) and characters
(synchronisation and block delimiters) are
needed to implement asynchronous and
synchronous transmission.
– These add nothing to the content of the message, but
must be included in what is sent.
– They reduce the overall information capacity of the
– They reduce the overall efficiency of the transmission.
Transmission efficiency
• Transmission efficiency = (useful data/total bits
– For example for asynchronous transmission of 8-bit
characters with 1 start and 1 stop bit, we have to send
10 bits for each character:
– Transmission efficiency = (8/10)*100 = 80%
• Effective Data Rate = (Transmission
Transmission Codes
• Symbolic data/information must be encoded in a
format suitable for transmission.
• Normally, the codes used for transmission are
similar to the codes used to store the
• The most common code is ASCII
– ASCII is a 7-bit code, permitting 128 different symbols to
be encoded.
• The second most commonly used code is EBCDIC
– EBCDIC is an 8-bit code enabling 256 different symbols
to be encoded.
Data Communication vs. Networking
• Communication: Two Nodes. Mostly EE issues.
Networking: Two or more nodes. More issues, e.g.,
Distributed Systems vs. Networks
Distributed Systems:
1. Users are unaware of underlying structure.
2. Mostly operating systems issues.
3. Nodes are generally under one organization’s
1. Users specify the location of resources.
2. Nodes are autonomous.
• Point to point communication not usually
– Devices are too far apart
– Large set of devices would need impractical number
of connections
• Solution is a communications network
What are computer networks?
Networking deals with the technology &
architecture of the communications networks
used to interconnect communicating devices.
Computer network is a collection of autonomous
computers interconnected by a single
The Internet is not a single network but a network
of networks.
Types of Networks
• Point to point vs. Broadcast
• Circuit switched vs. packet switched
• Local Area Networks (LAN)
Metropolitan Area Networks (MAN)
Wide Area Networks (WAN)
Communications Networks
• A Communications Network is a set of
interconnected devices that provide data
transmission facilities between user's end points.
User on
Host A
User on
Host B
Communications Network
comprising communications
links and connecting computers
Simplified Network Model
Objectives of Networking
• To share and exchange data between systems;
• To share expensive resources;
• To facilitate communication among humans and
Some terminology
• Host – a machine on the network;
• End system/end point – a machine “on the
edge” of the network, rather than an “internal”
(switching) node;
• Subnet – sub network, a subset of the whole
– Also used to refer to the internal “routing” part of a
• IMP – Interface Message Processor, hardware
connecting host to network.
Some terminology
• Packet – we often break messages into many
chunks, sent separately. The chunks are called
– Size of packet and how it’s treated depends on
network protocol in use.
– A packet might get split up further by another
– Some protocols (e.g. IP) use varying size packets; in
others (e.g. ATM) they’re fixed. Small fixed-size
packets are called cells
Some terminology
• internetworking – act of connecting multiple
networks together to form a larger network;
– Fun issues include how to route and address
across multiple heterogeneous networks;
• internet – a network thus produced
– Also the name of a common protocol for doing this
• Internet – “the” global internet;
Network sizes
• Computer Networks can be classified by the
area they cover:
PAN – Personal Area Network: very small
LAN – Local Area Network: room/building/campus
MAN – Metropolitan Area Network: city, region
WAN – Wide Area Network: country/continent.
Interconnection of Networks
• Networks of low capacity may be connected
together via a backbone (network of high
• LANs and WANs can be interconnected via T1
or T3 digital leased lines
• According to the protocols involved, networks
interconnection is achieved using one or several
of the following devices:
– Bridge: a computer or device that links two similar LANs based on the same protocol.
– Router: a communication computer that connects different types of networks using
different protocols.
– B-router or Bridge/Router: a single device that combines both the functions
of bridge and router.
– Gateway: a network device that connects two different systems, using direct and
systematic translation between protocols.
Broadcast vs. Point-to-point
• Broadcast Networks:
– A single communication channel shared by all
machines on a network;
• Multicast: simultaneous transmission to a subset.
• Point-to-point networks:
– Many connections between individual pairs of
– Transmission from A to C might go via B;
– Often multiple routes: a fundamental question is
which to use?
Local Area Networks
• A Local Area Network (LAN) is a computer
network intended to link computers and
associated devices within a small geographical
• The linking distances are relatively short, with cable lengths
rarely exceeding 5 kilometres.
• The linked computers may include large
computers, word processors, or desktop
• Associated devices include computer terminals, printers,
plotters, scanners, etc.
Local Area Networks
• LANs normally offer much higher data
transmission rates than WANs.
– This difference is apparent in the network oriented
protocols only.
• At application level, LANs provide the sharing of
resources like programs, files, printers, plotters,
scanners, etc.
LAN Topologies
• LAN topology is one of the issues that must be
considered when selecting LAN technology.
– It defines the interconnection of stations to form the
• LAN topologies are classified as:
– Broadcast topology
– Store-and-forward topology
LAN Topologies
• Broadcast topology
» This implies that all stations are connected
to a common transmission medium.
• Store-and-forward topology
» A complete message or packet is received
into a buffer in the memory of an
intermediate station
» It is then re-transmitted on the route to its
» The stations in a store-and-forward topology
network are connected by independent
point-to-point transmission lines.
LAN Topologies
• The topology of a LAN is important because it
influences the following features of the network:
– expansion cost
• the incremental cost of adding another station to an existing
– reconfiguration capabilities
• the ease of modifying the topology to deal with a failed node
or component.
– reliability
• The extent of dependency on a single component for network
LAN Topologies
• As well as:
• software complexity
– the complexity of the protocols required to achieve
• performance
– The effectiveness of the LAN in terms of throughput, or
delays in transmission.
• broadcast capabilities
– how difficult it is to broadcast in the LAN, i.e. to transmit a
single message which is received by all other stations in
the network.
Bus Topology
Single medium to which all
hosts are connected
Network connection
cable between NIC
on host and the
network medium
The bus is a broadcast topology –
anything placed on the medium by
one host is available to all hosts
attached to the bus.
If the destination address is
recognised by a node it copies the
contents into that node.
Ring Topology
The ring is a store-&-forward
topology – a packet placed on
the medium by one node is
received by the next node
along, and is then resent by
that node, and so on.
If the destination address
matches that of the receiver,
the packet is not resent.
If the sender’s address
matches that of the receiver,
the packet has gone all the way
round and will not be resent. It
is a “lost” packet and is
effectively removed from the
Star Topology
The star is a store-&-forward
topology – a packet placed on the
medium by one node is received
by central node which is forwarded
to it’s destination by the central
Central node
Hub Topology
• The “hub” is derivative of the bus and ring
– It has the appearance of the star topology, with a central
hub in place of the central node.
– The hub is simply the bus or ring wiring “collapsed” into a
central unit.
– Unlike the central node in the star topology, the hub does
not perform any switching. The hub simply consists of a
set of repeaters.
• Many modern networks are implemented using
hubs for convenience.
– Care is needed when deciding what topology is being
used in a real network.
Hub Topology :
Network with and without hub
Bus network – long network “backbone” installed through building, with short network
connection cables between host computers and backbone.
The same
bus network,
using a hub
Hub containing
“backbone” and
stubs of
cables extending
through the
building to the
host devices
Network topologies
• Tree
• Corresponding to
an organisational
• Internal nodes may
be bottlenecks.
Network topologies
• Graph
– Generalisation of a
– Cycles allowed
• Complete graph
– Dedicated link from
every node to every
other node
– Rapidly becomes
prohibitively expensive
Communications System
Communications System
• A communications system is the combination of
network hardware and communications system
software that supports the communications
between user-oriented processes running in
remote computers.
• The communications system provides the
services required by the applications to
communicate. These services are outlined on
the next slide.
Communications System
• Communication System Functions
– Naming and Addressing of entities.
– Segmenting and reassembly of messages
– Blocking of messages
– Connection or session control
– Error control
– Congestion and flow control
– Synchronisation
– Priority
Communication System Architecture
• The user-oriented layers:
• The application offers services to users through a
set of rules or steps for accessing web-sites or
sending e-mails.
– Some applications operate on different types of userinterface. A means of converting alphabets and screen
formats may be needed
– Some applications require a session of activity with a
definite set-up and closedown of the session (e.g. logon
and logoff)
• The transport layer provides an end-to-end
virtual channel between the source and
Communication System Architecture
• The system-oriented layers:
– Implement the connections between nodes that make
a machine part of a communications network
– The network layer is responsible for routing between
– The Data link and Physical layers provide the means of
moving packages of data between pairs of nodes.
Communication System Architecture
• The ISO Open Systems Interconnection
(OSI) model has 7 layers:
– The top 3 layers are user or application
– The bottom 3 layers are system-oriented.
– The middle layer, transport, acts as a
broker between the basic services provided
by the network and the needs of the users
• Each layer can be thought of as
“talking” directly to its peer on another
– A user of a web-browser holds a
“conversation” with a remote web-site
– Only at the physical layer does direct
communication take place, using signals.
1. Application
2. Presentation
3. Session
4. Transport
5. Network
6. Data Link
7. Physical
Communication System Architecture
• The TCP/IP model has 4 layers:
– The top layers is the application.
– The bottom 2 layers are system-oriented.
– The middle layer, transport, acts as a broker
between the basic services provided by the
network and the needs of the users.
• Although the model is simpler than OSI
it recognises the same purpose and
– The transport level protocols are TCP and
– The network level protocol is usually IP
– The data link and physical level protocols are
specific to the network
1. Application
2. Transport
3. Network
4. Data Link /
Communication System Architecture
1. Application
2. Transport
3. Network
4. Data Link /
Application-level protocol based on messages
used by the application. For example, the PIN
and card details from a bank card
Transport level data packets exchanged using
transport protocol
Network level data packets exchanged using
network protocol
Data Link frames exchanged using data link protocol.
Frames transmitted over the physical medium
using appropriate signalling techniques
1. Application
2. Transport
3. Network
4. Data Link /
Communication System Architecture
Host A
Host B
Intermediate nodes
inside network
1. Application
2. Transport
3. Network
4. Data Link /
Host A
1. Application
Application-level protocol
2. Transport
Transport-level protocol acting end-to-end – e.g. TCP
3. Network
4. Data Link /
Node C
3. Network
4. Data Link /
Node D
3. Network
4. Data Link /
Host B
Circuit Switching vs. Packet Switching
• Fundamental question: how to move bits from
one host to another, via ‘n’ others?
• Two key approaches (opposed):
– Circuit switching
• Establish fixed-bandwidth circuit & use it;
– Packet switching
• Split messages into packets, send separately;
• Trend is very much towards packet switching.
Circuit Switching
• Resources along a path are reserved for
duration of communication.
– Buffers, link bandwidth, CPU time, etc.
• All nodes on path genuinely maintain connection
state information;
• All data in a some communication is sent on the
same circuit, through same nodes;
• Classic example: PSTN (Public Switched
Telephone Network)
Circuit Switching
Circuit Switching
• Each circuit has a fixed bandwidth for its lifetime.
• Channels typically split into n equal bandwidth
• Pro: Makes QoS (Quality of Service) guarantees
easy to achieve;
• Con: Wasteful during “silent” periods.
– Data transmission tends to be bursty.
Circuit Switched Multiplexing
• Multiplexing – combining information channels
onto a common transmission medium.
• FDM (Frequency Division Multiplexing)
– Frequency spectrum of link is shared among circuits;
– Typically, each of n circuits gets 1/n;
– e.g. PSTN bandwidth divided in 4KHz bands;
• TDM (Time Division Multiplexing)
– Time divided into fixed size chunks;
– Each circuit gets a portion of the total time
Packet Switching
• No prior reservation of resources;
• Each packet transmitted separately;
• Nodes don’t maintain connection state
– Each packet dealt with individually;
– Two packets might take different paths;
• Classic example: the Internet.
Packet Switching
• Con: QoS harder to do, can only really make
“best effort” promises;
– IPv6 addresses this somewhat – complex;
• Pros: more efficient use of bandwidth, no hard
limit to number of comms.
– Ideally: “graceful degradation” curves;
– What happens when queues fill? Delays and,
ultimately, packet loss.
• Store-and-forward (on routers):
– Read entire packet in, then send it out
Packet switching
Delay & Loss in Packet Switching
• Processing delay
– Time to examine packet & decide where to send it;
maybe also some error checking;
• Queuing delay
– Delay while packet is queued; depends on size of
queue, ie traffic levels;
• Transmission delay
– Time taken for node to “push out” packet;
– Depends on size of packet & speed of outbound link.
Delay & Loss in Packet Switching
• Propagation delay
– Time taken for packet to propagate across link to next
– Depends on speed of physical medium and distance
to next node;
• Packet loss
– Happens when things get too busy, queues overflow,
nodes can’t keep up;
• End-to-end delay
– Total delay on transmission between two end points.
Frame Relay
• Packet switching systems have large overheads
to compensate for errors
• Modern systems are more reliable
• Errors can be caught in the end system
• Most overhead for error control is stripped out
Asynchronous Transfer Mode
Evolution of frame relay
Little overhead for error control
Fixed packet (called cell) length
Anything from 10Mbps to Gbps
Constant data rate using packet switching
Virtual circuits vs. datagram networks
• We can, in fact, simulate circuit switching on
packet switched networks:
– Virtual Circuits being the result;
• Otherwise, it’s a datagram network:
– Datagram: another word for packet;
• Choice has huge impact on routing;
– At IP level, Internet is a datagram network
Virtual Circuit Networks
• Packets carry VC identifier;
• Hosts have table mapping VCIDs to outbound
• Setting up involves both ends and every host in
• Every packet follows the same path;
• Requires complex state maintenance protocols.
Datagram Networks
• Packets carry destination address;
• Host has (more complex) table to help it decide
where to send next.
– Table at a given host can change over lifetime of a
• Packets really can take different paths;
• No connection state information maintained
(except maybe at ends);
• Almost all of the Internet.
Connection-oriented vs. Connectionless
• Characterises end-to-end communication
services available to end users.
• Connection-oriented:
– Application must establish connection to other end
before sending any actual data;
– Each packet then sent via that connection.
• Allows delivery guarantees.
• Connectionless:
– Application just sends each packet individually;
– Thus, must know destination address every time you
send a packet.
• No guarantee of delivery, generally.
Caution: don’t get confused…
• Circuit-switched vs. packet switched:
• Concerns how packets are routed;
• Distinction made in “core” of network;
• Mainly at Network layer (see later).
• Connection-oriented vs. connectionless:
• Concerns how packets are sent/received;
• Distinction made at “edge” of network;
• Mainly at Transport layer (see later).
Basic Types of Networks
Yet another way to classify…
Basic Types
Does not require dedicated resource (dedicated server)
Any host can share its resources
Typically less expensive, easier to work with
Less secure, support fewer users (10 or fewer), experience more
problems with file system management
Configuration of nodes, certain of which are dedicated to
providing resources (servers)
Offer (better) user security
Dedicated servers can be expensive, may require a full-time
network administrator
Enterprise network (which combines the two)
Provide connectivity among all nodes in an organization
Include (connect) both peer-to-peer and server-based
May consist of multiple protocol stacks and network
Client/Server Networks
• Client/Server is a networking model mainly
applicable at the Application layer;
• Concerns the roles of end systems:
• Client – system requesting some service;
• Server – system providing some service.
• Ubiquitous example: HTTP
• Client is your web browser
• Server is (or whatever)
Peer Networks
• Not all applications use Client/Server model;
• Often, all parties have equal status:
– In some sense they’re all clients and servers.
– Although sometimes have distinguished nodes
providing certain services.
• Used for communications between entities in a
• Must speak the same language
• Entities
– User applications
– e-mail facilities
– terminals
• Systems
– Computer
– Terminal
– Remote sensor
Key Elements of a Protocol
• Syntax
– Data formats
– Signal levels
• Semantics
– Control information
– Error handling
• Timing
– Speed matching
– Sequencing
Protocols define format, order of messages sent
and received among network entities, and
actions taken on message’s transmission,
In Summary, a protocol is ....
• An agreement about communication between
two or more entities
• It specifies
– Format of messages
– Meaning of messages
– Rules for exchange
– Procedures for handling problems
Protocol Architecture
• Task of communication broken up into modules
• For example file transfer could use three
• File transfer application
• Communication service module
• Network access module
Simplified File Transfer Architecture
A Three Layer Model
• At the Top:
– User Oriented layer-Application Layer
• In the Middle:
– Transport Layer
• At the Bottom:
– System Oriented Layer - Network Access Layer
Network Access Layer
• Exchange of data between the computer and the
• Sending computer provides address of
• May invoke levels of service
• Dependent on type of network used (LAN,
packet switched etc.)
Transport Layer
• Reliable data exchange
• Independent of network being used
• Independent of application
Application Layer
• Support for different user applications
• e.g. e-mail, file transfer
Addressing Requirements
• Two levels of addressing required
• Each computer needs unique network address
• Each application on a (multi-tasking) computer
needs a unique address within the computer
• The service access point or SAP
• The port on TCP/IP stacks
• Different levels of entity use different addresses.
• MAC address: Identifies the NIC and set by
– Used by Physical and Data Link layer
• IP address: Identifies a computer in a network.
– Used by the Network layer
• Socket: Identifies a process (running program).
– Used by the Transport layer
– Application level addresses vary:
• One example is the Uniform Resource Locator
(URL) used by WWW applications
IP Addresses
• IP = Internet Protocol
• Each IP address is 32 bits long
• An IP address has a network part and host part
• The former identifies a specific network and the latter a
specific computer, or host, on that network.
• IP addresses may be in one of five network
– Class A: Used for a small number of networks, each with many
– Class B: Used for a larger number of networks, each with a
medium number of hosts
– Class C: Used for a large number of networks, each with only a
few hosts
– Classes D and E are for special purposes.
IP Addressing Example
• All hosts on a network have the same network
User Oriented Names and DNS
• Human users prefer names to numbers.
• The communications system translates these
names into IP addresses, and vice versa.
• The translation is done using the Domain Name
System (DNS) application.
– This is a “directory” service.
– It uses multiple levels of server to resolve
queries as close to the point of issue as
– All servers cache query results to reduce need
for repeat queries in the near future.
Name Resolution in DNS
• Each computer has a name resolver routine
– ‘gethostbyname’ in UNIX
• Each resolver knows the name of a local DNS
• Resolver sends a DNS request to the server
• DNS server either gives the answer, forwards
the request to another server, or gives a referral
– Referral = Next server to whom request should be
How the DNS works
Intermediate Name Server
Root Name Server
Name embedded
in query
Authoritative Name Server
IP address, or error
message, embedded
in reply
Local Name Server
Requesting Host
A “requesting host” issues a query to its local name
server to translate from a host name to an IP address.
This request is routed to a server which can supply the
IP address.
The root server may not know the IP address, but may
“know the address of someone who does”. This can
result in queries being forwarded to an “authoritative
name server, via an intermediate name server, if
All name servers employ caches which store the results
of recent queries. This can speed up later requests for
the same IP address.
Protocol Architectures and Networks
Protocols in Simplified Architecture
Protocol Data Units (PDU)
• At each layer, protocols are used to
• Control information is added to user data at
each layer
• Transport layer may fragment user data
• Each fragment has a transport header added
– Destination SAP
– Sequence number
– Error detection code
• This gives a transport protocol data unit
Protocol Data Units
Network PDU
• Adds network header
• network address for destination computer
• Facilities requests
Operation of a Protocol Architecture
• Required to allow for interoperability between
• Advantages
• Ensures a large market for equipment and
• Allows products from different vendors to
• Disadvantages
• Freeze technology
• May be multiple standards for the same thing
Standardized Protocol Architectures
• Required for devices to communicate
• Vendors have more marketable products
• Customers can insist on standards based
• Two standards:
• OSI Reference model
– Never lived up to early promises
• TCP/IP protocol suite
– Most widely used
• Open Systems Interconnection
• Developed by the International Organization for
Standardization (ISO)
• Seven layers
• A theoretical system delivered too late!
• TCP/IP is the de facto standard
OSI - The Model
• A layer model
• Each layer performs a subset of the required
communication functions
• Each layer relies on the next lower layer to
perform more primitive functions
• Each layer provides services to the next higher
• Changes in one layer should not require
changes in other layers
OSI Layers
The OSI Environment
TCP/IP Protocol Architecture
• Developed by the US Defense Advanced
Research Project Agency (DARPA) for its packet
switched network (ARPANET)
• Used by the global Internet
• Not official model but a working one.
Application layer
Host to host or transport layer
Internet layer
Network access layer
Physical layer
TCP/IP Protocol Architecture:
Physical Layer
• Physical interface between data transmission
device (e.g. computer) and transmission
medium or network
• Characteristics of transmission medium
• Signal levels
• Data rates
• etc.
TCP/IP Protocol Architecture:
Network Access Layer
• Exchange of data between end system and
• Destination address provision
• Invoking services like priority
TCP/IP Protocol Architecture:
Internet Layer (IP)
• Systems may be attached to different networks
• Routing functions across multiple networks
• Implemented in end systems and routers
TCP/IP Protocol Architecture:
Transport Layer (TCP)
• Reliable delivery of data
• Ordering of delivery
TCP/IP Protocol Architecture:
Application Layer
• Support for user applications
• e.g. http
TCP/IP Protocol Architecture Model
Protocol Data Units in TCP/IP