Internetworking with TCP/IP

CoSc 611
Credit Hours : 3 Lecture Hours
Year: I Semester: I

[Link] Periyasamy
 Origins of TCP/IP
 1950’s – 1960’s – US Govt. requirement for “rugged(uneven surface)”
network that would continue to work in case of a nuclear attack

 RAND Corporation (America’s leading think thank) & DoD formed ARPA
(Advanced Research Project Agency)

 1968 – ARPA engineers proposed Distributed network design for

ARPANET Network
 ARPANET - an early packet switching network and first network to implement the
protocol suite TCP/IP. Both technologies became the technical foundation of the
Internet
Distributed Network Design
 Pre-ARPANET networks

– “connection oriented”

– Management & control was centralized

 “New” Network – ARPANET

– Connectionless

– Decentralised

 Modern Internet has evolved from the ARPANET

 A brief history of the Internet
1961-1972: Early packet-switching principles

 1961: Kleinrock - queuing theory  1972:

shows effectiveness of packet- – ARPAnet public demonstration
switching – NCP (Network Control Protocol) first host-
host protocol
 1964: Baran - packet-switching in – first e-mail program
military nets – ARPAnet has 15 nodes

 1967: ARPAnet visualized by

Advanced Research Projects
Agency

 1969: first ARPAnet node

operational
 A brief history of the Internet
1972-1980: Internetworking, new and proprietary nets

 1970: ALOHAnet satellite network in

Cerf and Kahn’s internetworking
Hawaii
principles:
 1974: Cerf and Kahn - architecture for
– minimalism, autonomy - no internal
interconnecting networks
changes required to interconnect
 1976: Ethernet at Xerox PARC
networks
 late70’s: proprietary architectures:
– best effort service model
DECnet, SNA, XNA

 late 70’s: switching fixed length packets

– stateless routers

(ATM precursor) – decentralized control

 1979: ARPAnet has 200 nodes
 A brief history of the Internet
1980-1990: new protocols, a proliferation of networks

 1983: operation of TCP/IP  New national networks: Csnet,

 1982: SMTP e-mail protocol defined BITnet, NSFnet, Minitel

 1983: DNS defined for name-to-IP-  100,000 hosts connected to

address translation confederation of networks

 1985: FTP protocol defined

 1988: TCP congestion control

 A brief history of the Internet
1990, 2000’s: commercialization, the Web, new apps

 Early 1990’s: ARPAnet decommissioned  Late 1990’s – 2000’s:

 1991: NSF lifts restrictions on – more killer apps: instant messaging,
commercial use of NSFnet P2P file sharing
(decommissioned, 1995) – network security to forefront

 Early 1990s: Web – est. 50 million host, 100 million+

users
– hypertext [Bush 1945, Nelson 1960’s]
– backbone links running at Gbps
– HTML, HTTP: Berners-Lee

– 1994: Mosaic, later Netscape

– late 1990’s: commercialization of the

Web
Simplified view of the Internet
 Internet works
 Start with lots of little networks and many different types

– Ethernet, dedicated leased lines, dialup, ATM, Frame Relay, FDDI

 Each type has its own idea of addressing and protocols

 Want to connect them all together and provide a unified view of the whole lot (i.e.
act as a single large network)
Network Connection Types
 The “nuts and bolts” view of the Internet

 millions of connected computing

router
devices: hosts, end-systems workstation
– PCs workstations, servers server
mobile
– PDAs phones
local ISP
running network apps
 communication links
– fiber, copper, radio, satellite regional ISP
– Links have different bandwidth
 routers: forward packets
 Packet: a piece of messg.
company
network 1-11
 How should network allocate resources

– circuit switching: dedicated

circuit per call: telephone
network

– packet-switching: data sent

through network in discrete
“chunks”
 Circuit Switching

 End-to-end resources (eg

link bandwidth, switch
capacity) reserved for “call”

 Dedicated allocation, ie, no

sharing

 Guaranteed performance

 Call setup required

 Circuit Switching -FDM and TDM
Dividing link bandwidth across calls
Example:
FDM
4 users

frequency

time
Each mobile’s share of the bandwidth is divided into portions for the uplink and the downlink.

TDM
frequency

time
 Packet Switching
Each end-end data stream divided Resource conflict:
into packets
 Collective resource demand can
 Packets across flows share network
resources exceed than amount available

 Congestion: packets queue, wait for

 Each packet uses full link bandwidth link use

 Store and forward

 Resources used as needed
 packets move one hop at a time

 node receives complete packet

Dedicated allocation before forwarding
Resource reservation
 Packet switching versus circuit switching

Packet switching allows more users to use network!

 1 Mb/s link
 Each user:
– 100 kb/s when “active”
– active 10% of time
N users
 Circuit-switching: 1 Mbps link
– 10 users
 Packet switching:
– with 35 users, probability > 10
active less than .0004
 Edge vs. core functionality

 Telephone networks: “dumb” end-systems, complex

network

 Internet: “smart” end-systems, simple network

 Design philosophy of the DARPA Internet
[Clark88]

 Top level goal: “to develop an effective technique for

multiplexed utilization of existing interconnected

networks”
 Seven second level goals

1. Internet communication must continue despite loss of

networks or gateways.
 Assumptions/implications at end-host
– (no connection setup, route diversity exists, adaptive
network-layer, end-host oblivious to network information)
– (stateless)
– (fate sharing) an engineering design philosophy where related parts of a
system are yoked together, so that they either fail together or not at all.
Fate-sharing is an example of the end-to-end principle
 Seven second level goals
2. Internet must support multiple types of communications service
 Layering implications
– TCP maybe inappropriate (eg, ping) or overkill (eg, real-time audio/video), so
TCP must be in a separate layer from IP
– TCP and UDP and many other protocols co-exist on top of IP

[Link] must accommodate a variety of networks

4. Internet must permit distributed management of resources

5. Efficiently use resources

6. Permit low-effort host attachment

7. Enable accountable resource usage

 What is TCP/IP?

 In simple terms is a language that enables communication between

computers

 A set of rules (protocol) that defines how two computers address

each other and send data to each other

 Is a suite of protocols named after the two most important protocols

TCP and IP but includes other protocols such as UDP, RTP, etc
 Open Systems & TCP/IP
TCP/IP formed from standardized communications procedures that were
platform independent and open

Open systems

– open architecture - readily available to all

What is open system networking?

– network based on well known and standardized protocols

– standards readily available

– networking open systems using a network protocol

 To install TCP/IP protocol support - follow steps:

1. Open the Network Settings dialog box (double-click the Network icon in the Control Panel).

2. Click Add in the Protocols tab to open the Select Network Protocol dialog box.

3. Select TCP/IP Protocol in the Network Protocol list and choose OK.

4. The next prompt asks, “Do you wish to use DHCP?” If this computer will obtain its IP
address from DHCP, choose Yes. If this computer will be configured with a static IP
address, choose No.

5. When prompted, supply the path where Setup can locate the driver files.

6. Choose Close to exit the Network settings dialog box. After recalculating the bindings, Setup
shows you a Microsoft TCP/IP Properties dialog box that will, at first, be blank.
 Contd….

7. If more than one adapter has been installed, select the adapter to be configured in the
Adapter list. (You should configure each adapter with a valid IP address for the subnet they
are on.)
[Link] this computer will obtain its address configuration from DHCP for any of the network
adapters, click the Obtain an IP address from a DHCP server radio button.
9. If this computer will be configured with static addresses, click the Specify an IP address
radio button and complete the following fields:
IP Address (Required)
Subnet Mask (Required. Setup will suggest the default subnet mask appropriate for the IP
address you enter.) Default Gateway
10. Choose OK and restart the computer to activate the settings.
 TCP/IP Services
 Line Printer Daemon - enables you to share printers with many different types of hosts.

 DHCP - provides automatic configuration of remote hosts, making management of a TCP/IP

environment easy.

 DHCP Relay Agent - extends the capabilities of DHCP service by allowing it to work across various
different subnets.

 Windows Internet Name Service (WINS) - WINS server provides a centralized method of name
management that is both flexible and dynamic.

 Simple Network Management Protocol Agent (SNMP) - track the performance of your TCP/IP
protocols.

 Domain Name Server (DNS) - works with host names to allow you to integrate your systems into the
Internet or to resolve hosts on the Internet.
 Topology
 The physical topology of a network refers to the configuration of cables,

computers and other peripherals.

 The main types of network topologies are:

– Linear Bus

– Star

– Ring

– Mesh

– Tree or Hybrid
 Linear Bus topology
 A linear bus topology consists of a main run of cable with a terminator at
each end.

 All servers workstations and peripherals are connected to the linear cable
 Advantages and Disadvantage of Bus topology

1) Easy to implement 1) Difficult to administer/troubleshoot

2) Limited cable length and number of

2) Well suited for temporary networks that
stations
must be set up in a hurry 3) A cable break can disable the entire
network;
3) Typically the least cheapest topology to
4) Maintenance costs may be higher in the
implement
long run

5) Performance degrades as additional

computers are added

28
 Star topology
 A star network is designed with each node (file server, workstation,

peripheral) connected directly to a central network hub or server

 Advantages and Disadvantages of star topology

1) If central device fails whole network

1) Compared to Bus topology it gives far
goes down
much better performance
2) The use of hub, a router or a switch as

2) Easy to connect new nodes or devices central device increases the overall

cost of the network

3) Centralized management. It helps in
3) Performance and number of nodes
monitoring the network
which can be added is dependent on

capacity of central device

30
 Ring topology

 A ring network is one where all workstations and other devices are

connected in a continuous loop. There is no central server

 Advantages and Disadvantage of Ring topology

1) Well organized 1) Each packet of data must pass through all

2) Performance is better than Bus topology the computers between source and

3) No need for network server to control the destination, slower than star topology

connectivity between workstations 2) If one workstation or port goes down, the

4) Additional components do not affect the entire network gets affected

performance of network 3) Network is highly dependent on the wire

5) Each computer has equal access to resources which connects different components

32
 Advantage and disadvantage of Mesh Topology

 ▸ possible to transmit data from one [Link] leads to idleness of many network
node to many other nodes at the same
time. connections.


▸ The failure of a single node does not [Link] incurred in setup and maintenance are
cause the entire network to fail
high.

▸ It can handle heavy traffic
 3. Owing to its complexity, the administration
▸ Point-to-point contact between every
pair of nodes, makes easy to identify of a mesh network is difficult.
faults.

34
 Tree or hybrid topology
 A tree or hybrid topology combines characteristics of linear bus and star
and/or ring topologies.

 It consists of groups of star-configured workstations connected to a linear

bus backbone cable
 Network Operating Software

 Network operating systems co-ordinate the activities of multiple computers

across a network

 The two major types of network OS are:

– Peer-to-peer

– Client/server
 Peer to peer network OS
– In P2P network OS, there is no file server or central management source; all
computers are considered equal

– Peer to peer networks are designed primarily for small to medium LANS

– Ex - AppleShare and Windows for Workgroups

 Client/Server network OS
– Client/server network OS centralize functions and applications in one or more
dedicated file servers.

– The file server provides access to resources and provides security

– Ex - Novel Netware and Windows NT Server

 OSI - Layered Model Concept
 Open Systems Interconnection Reference Model (OSI-RM) adopted as a
standard for networking

 Divide-and-conquer approach

 Dividing requirements into groups, e.g transporting of data, packaging of

messages, end user applications

 Each group can be referred to as a layer

– Upper layers are logically closer to the user and deal with more abstract data

– Lower layer protocols translate data into forms that can eventually be
physically transmitted.
OSI Model
 OSI Model

7 Application
APPLICATION
6 Presentation
• Upper Layers
Session • Application oriented
5
• Independent of layers below
4 Transport

3 Network
TRANSPORT
Data Link • Lower Layers
2
• Transmission of data
1 Physical • No differentiation of upper layers
 Layers 7, 6, 5
 7: Application layer

– Provides different services to the applications

– Uses the underlying layers to carry out work

» e.g. SMTP (mail), HTTP (web), Telnet, FTP, DNS

 6: Presentation layer

– Converts data from applications into common format and vice versa

 5: Session layer

– organizes and synchronizes the exchange of data between application

processes
 Layer 4
 4: Transport layer

– Provides end to end transportation of segments

– E.g. TCP

» sum up TCP segments in network layer packets

» adds reliability by detecting and retransmitting lost packets

» uses acknowledgements and sequence numbers to keep track of

successful, out-of-order, and lost packets

» timers help differentiate between loss and delay

– UDP is much simpler: no reliability features

 Layer 3
 3: Network layer
– Routes the information in the network

– E.g. IP is a network layer implementation which defines

addresses in such a way that route selection can be determined.
» Single address space for the entire internetwork

» adds an additional layer of addressing, e.g. IP address, which is

different from MAC address.
 Layer 3
 3: Network layer (e.g. IP)
– Unreliable
» if packet gets lost, network layer doesn’t care for higher layers can resend lost
packets

– Forwards packets hop by hop

» encapsulates network layer packet inside data link layer frame

» different framing on different underlying network types

» receive from one link, forward to another link

» There can be many hops from source to destination

 Layer 3
 3: Network layer (e.g. IP)

– Makes routing decisions

» how can the packet be sent closer to its destination?

» forwarding and routing tables represent “knowledge” of network

topology

» routers can talk to each other to exchange information about network

topology
 Layer 2
 2: Data Link layer

– Provides reliable transit of data across a physical network link

– bundles bits into frames and moves frames between hosts on the same

link

– a frame has a definite start, end, size

– some link layers detect corrupted frames while other layers re-send

corrupted frames (NOT Ethernet)

 Layer 1

 1: Physical layer

– moves bits using voltage, light, radio, etc.

– no concept of bytes or frames

– bits are defined by voltage levels, or similar physical properties

1101001000
 OSI and TCP/IP
7 Application
Mail, Web, etc.
6 Presentation Application
5 Session
4 TCP/UDP – end to end reliability
Transport Transport
3 Network Network IP - Forwarding (best-effort)

2 Data Link Data Link & Framing, delivery

1 Physical Physical Raw signal

OSI TCP/IP
Protocol Layers:
The TCP/IP Hourglass Model
Application layer
SMTP HTTP FTP Telnet DNS Audio Video

TCP UDP RTP Transport layer

IP Network layer

Token Frame
Ethernet ATM X.25 PPP HDLC
Ring Relay
Data link layer
 Layer Interaction
Application, Presentation and Session protocols are end-to-end

Transport protocol is end-to-end

– encapsulation/ decapsulation over network protocol on end systems

Network protocol is throughout the internetwork

– encapsulation/ decapsulation over data link protocol at each hop
– Link and physical layers may be different on each hop
 Layer Interaction:
OSI 7-Layer Model
Application Application

End Presentation Presentation

to Session Session
end Transport Transport
Network Network Network Network
Hop
Link Link Link Link Link Link
by
hop Physical Physical Physical

Host Router Router Host

 Layer Interaction:
TCP/IP Model
No session or presentation layers in TCP/IP model
End
to Application Application
end TCP or UDP TCP or UDP
IP IP IP IP
Hop
Link Link Link Link Link Link
by
hop Physical Physical Physical

Host Router Router Host

Encapsulation & Decapsulation
 Lower layers add headers (and sometimes
trailers) to data from higher layers
Application Data

Transport Header Transport Layer Data

Network Header Network Layer Data

Network Header Header Data

Data Link Header Link Layer Data Trailer

Data Link Header Header Header Data Trailer
 Purpose of an IP address
 Unique Identification of

– Source(used for security or policy-based filtering of data)

– Destination(So the networks know where to send the data)

 Network Independent Format

– IP over anything

 Identifies a machine’s connection to a network

 Physically moving a machine from one network to another requires changing

the IP address

 TCP/IP uses unique 32-bit addresses

 Basic Structure of an IP Address
 32 bit number (4 octet number) (e.g. [Link])
On a machine running Windows NT 4.0, it is relatively easy to determine its IP
address.
Complete the following steps:
1. From the Start menu, select Programs, Command Prompt.
2. After the command prompt window appears, type IPCONFIG /all.
3. Read the information provided by the IPCONFIG utility until you see a
section called “Ethernet address.”
- The value represented is the physical address of the machine.
 Addressing in Internetworks
 The problem we have

– More than one physical network

– Different Locations

– Larger number of computers

 Need structure in IP addresses

– network part identifies which network in the internetwork (e.g. the Internet)

– host part identifies host on that network

 Address Structure Revisited
 Hierarchical Division in IP Address:
– Network Part (Prefix)
» describes which physical network
– Host Part (Host Address)
» describes which host on that network

205 . 154 . 8 1
11001101 10011010 00001000 00000001
Network Host
– Boundary can be anywhere
» very often NOT at a multiple of 8 bits
 Classless Addressing
 IP address with the subnet mask defines the range of
addresses in the block
– E.g [Link]/28 (subnet mask [Link]) defines the
range [Link] to [Link]

– [Link] is the network address

– [Link] is the broadcast address

– [Link] ->46 assignable addresses