Prof Pallapa. Venkataram,

1
Network Reference Model

• Prof Pallapa. Venkataram,
• Electrical Communication Engineering,
• Indian Institute of Science,
• Bangalore 560012, India

2
What is Network Reference Model?

• A network is a group of computers, printers and
  other devices that are connected together either
  with a cable or wireless media.
• The network users share hardware or software over
  the network.
• A network reference model clearly defines the
  functions of communication softwares in a
  generalized and structured manner which helps to
  carry out the network product development
  activities.

3
Network Architecture
Abstract representation of network architecture
4
Network Reference Model
Layered architecture,
5
Features of Layered Architecture

• Collection of protocols
• network reference model.
• An entity could be a software (or a process) or
  hardware (circuit/chip) entity.
• The entities in the same layer on different
  machines are called as peer entities.
• succession of logically distinct entities
• 'n'th entity provides services to 'n1'th entity
  and gets service from 'n-1'th entity.

6
Peer to Peer Model
Interaction between two layers
7
Advantages of the Layered Model

• Explicit structure allows identification of the
  relationship among a complex system's pieces.
• Modularization eases maintenance and updating of
  the system.
• Change of implementation of a layer's service
  transparent to rest of system.
• Without layering, each new application has to be
  re-implemented for every network technology.

8
Network Services Interfaces

• Service
• A set of service primitives (operations) that a
  layer provides to the layer above it.
• A service relates to an interface between two
  layers, with the lower layer being the service
  provider and the upper layer being the service
  user.
• Interface
• a point at which the services are accessible to
  the layers.
• The services are available through the Service
  Access Points (SAPs).
• The layer 'n' SAPs are the places where layer
  'n1' can access the services offered.
• Each SAP has an address that uniquely identifies
  it.

• Fig. 2.3 Layer interface and the control units

9
Protocol Functions

• Protocol
• a set of rules governing the format and meaning
  of frames, packets, or messages that are
  exchanged by the peer entities within a layer.
• Functions of a Protocol
• Encapsulation, segmentation and reassembly of
  messages
• Connection control
• Ordered delivery
• Error control, Flow control Multiplexing
• Addressing

10
Encapsulation, Segmentation and Reassembly of
messages

• Encapsulation
• is the technique used by layered protocols in
  which a layer accepts a message from a layer
  above it and places it in the data portion of the
  lower level layers message unit.
• As data moves down the layers, additional
  information will be appended to it, and it may be
  segmented into smaller pieces.
• Segmentation
• Multiplexing and error control require messages
  to be of a maximum length, and application
  messages must be divided into segments that match
  the transmission criteria.
• Application messages that are divided must be
  reassembled before being presented to destination
  application.

11
Error control, flow control and multiplexing

• Error control is needed for reliable end-to-end
  (host-to-host) and link (network node-to-node)
  transmission.
• Forward Error Correction or ARQ.
• Link level error control
• Flow control

12
Multiplexing Addressing.

• Types of multiplexing
• 1. upward multiplexing, i.e., layer 'n'
  multiplexes the connections from the layer 'n1',
• 2. downward multiplexing, i.e., layer 'n1'
  multiplexes the connections from layer 'n'.
• Similarly, the layers demultiplex the connections
  at the destination side of the network.
• Addressing
• Destination Identification
• Addressing allows the separation of the
  application- application dialog from the
  selection process.
• Layers allow interim addresses for resilient
  routing while preserving global application
  addresses.

13
OSI Model
14
OSI Model Overview

• The upper layers (5-7) of the OSI model deal with
  application issues and generally are implemented
  only in software.
• The highest layer, the application layer, is
  closest to the end-user.
• Both users and application layer processes
  interact with software applications that contain
  a communications component.
• The lower layers (1-4) of the OSI model handle
  data transport issues.
• The physical layer and the data link layer are
  implemented in hardware and software.
• The lowest layer, the physical layer, is closest
  to the physical network medium (for example,
  network cabling) and is responsible for actually
  placing information on the medium.

15
Data transfer across the OSI layers
Where a message 'x' is transmitted from host A to
host B. At each layer, a header (AH-Application
Header, PH-Presentation Header, SH-Session
Header, TH-Transport Header, NH-Network Header,
DH-Data link Header) is attached to the message.
A header and trailer is attached at data link
layer. The trailer includes checksum bits for
error detection at link level.
16
Physical Layer

• Provides direct mechanical and electrical
  connections between the computer system and the
  network nodes.
• The physical layer has set of interfacing rules
  to communicate with devices like modems, the
  broadband, carrier-band or other modulation
  techniques that puts signals on the network.
• Physical layer specifications defines
  characteristics such as voltage levels, timing of
  voltage changes, physical data rates, maximum
  transmission distances, and physical connectors.

17
Data link layer

• establishes and maintains a communication path
  between nodes of the network.
• responsible for transferring frames from one
  computer to another, without errors.
• establishes connections upon request by the
  network layer and disconnects them after the
  completion of the transmission.
• perform flow control, which moderates the
  transmission of data so that the receiving device
  is not overwhelmed with more traffic than it can
  handle at one time.
• Data link layer is divided into two sub-layers
• Logical Link Control (LLC) manages
  communications between the devices over a single
  link of a network, and supports both
  connection-less and connection-oriented services
  used by upper layer protocols.
• Media Access Control (MAC) manages protocol
  access to the physical network medium. The MAC
  defines MAC addresses, which enable multiple
  devices to uniquely identify one another at the
  data link layer.

18
Network Layer Transport Layer

• NETWORK LAYER
• Concerned with routing of data from one network
  node to another.
• defines network layer implementations, network
  addresses in a way that route selection can be
  determined systematically by comparing the source
  network address with the destination network
  address and applying the subnet mask.
• It also has the responsibility of interconnecting
  two or more similar/dissimilar networks.
• TRANSPORT LAYER
• accepts the data from session layer and segments
  the data for transport across the network.
• responsible for making sure that the data is
  delivered error-free and in the proper sequence.
• decides to multiplex transport connections either
  upwards (from a single network connection to
  several transport connections) or downwards
  (splitting a single transport connection among
  many network connections).

19
Session layer Presentation Layer

• SESSION LAYER
• establishes, manages, and terminates
  communication sessions.
• Communication sessions consist of service
  requests and service responses that occur between
  applications located in different network
  devices.
• It determines which of the three modes of
  interaction the communication may take simplex,
  half-duplex, or full-duplex.
• It synchronizes and checks the points to keep the
  two end devices in step with each other.
• PRESENTATION LAYER
• Provides a variety of coding and conversion
  functions that are applied to application layer
  data.
• These functions ensure that information sent from
  the application layer of one system would be
  readable by the application layer of another
  system.
• Some data coding and conversion schemes used in
  presentation layer include common data
  representation formats, conversion of character
  representation formats, common data compression
  schemes, and common data encryption schemes.

20
Application layer

• The application layer is the OSI layer closest to
  the end user, which means that OSI application
  layer interacts with the user through the user
  interface programs.
• Its primary function is to provide the mechanisms
  and interfaces that enable an end user to
  communicate within the network environment.
• Functions may include logging in, checking
  password, file request, file transfer, etc.

21
TCP/IP protocol suite
22
TCP/IP protocol suite functioning
23
MAC Protocols

• used for sharing a single broadcast channel among
  several users by avoiding conflicts between the
  sending hosts.
• three types of MAC protocols
• Channel partitioning divides the channel into
  smaller pieces (in terms of either time slots or
  frequency) and allocates a piece to a node for
  exclusive use.
• eg. TDMA, FDMA and CDMA.
• TDMA (Time Division Multiple Access)
• access to a channel by the hosts is in rounds,
• i.e., each station gets a fixed length slot
  (length packet transmission time) in each
  round.
• The unused slots by nodes go idle and hence
  bandwidth is wasted.
• FDMA (Frequency Division Multiple Access)
• channel spectrum is divided into frequency bands
  where each host is assigned a xed frequency
  band.
• The unused transmission time in frequency bands
  go idle and hence bandwidth is wasted.

24
MAC Protocols

• CDMA (Code Division Multiple Access)
• a unique code is assigned to each host in a
  network.
• used in wireless broadcast channels (cellular,
  satellite, etc.) where all hosts share same
  frequency, but each host has its own chipping
  sequence (i.e., code) to encode the data.
• The encoded signal is given as product of
  original data and chipping sequence.
• The decoding inner-product of encoded signal and
  chipping sequence allows multiple hosts to
  coexist and transmit simultaneously with minimal
  interference (if codes are orthogonal).
• Random Access protocols allows collisions
  between the nodes, and uses some mechanisms to
  recover from collisions
• eg. pure Aloha, slotted Aloha, CSMA/CD, etc.
• pure Aloha
• the nodes can start transmission as and when the
  frames are ready. If transmission is not
  successful, a node retransmits the frame.
• This protocol has higher collision probability.

25
MAC Protocols

• Slotted Aloha
• considers all frames of same size since it
  divides time into equal sized slots which is
  given as the time to transmit one frame.
• The nodes start transmitting frames only at
  beginning of a slot (nodes are synchronized).
• In case if two or more nodes transmit in a slot,
  all nodes will detect the collision and try
  retransmitting in each subsequent slot with
  probability 'p' until success.

26
MAC Protocols

• CSMA (Carrier Sense Multiple Access)
• if a node senses the channel to be idle, it
  transmits an entire frame, otherwise defers the
  transmission.
• CSMA/CD (Carrier Sense Multiple Access/ Collision
  Detection)
• same as in CSMA but collisions are detected
  within a short time, and the colliding
  transmissions are aborted and scheduled after
  some random time, thus reducing channel bandwidth
  wastage.
• The MAC protocols IEEE Standards 802.3 (CSMA/CD),
  802.4 (token bus), 802.5 (token ring), and 802.11
  (wireless CSMA/CA, i.e., Carrier Sense Multiple
  Access with Collision Avoidance) are used for
  accessing media in wired bus, wired logical ring,
  wired ring and wireless network topologies,
  respectively.

27
MAC Protocols

• Taking Turn protocols share the channel either
  by polling or token passing methods.
• Polling
• a master node invites the slave nodes to transmit
  in turn.
• Issues polling overhead, latency, and single
  point of failure (master).
• Token Passing method
• a control token is passed from one node to next
  sequentially where presence of token indicates
  transmit permission.
• issues token overhead, latency, and single point
  of failure (token).

28
LLC Protocols

• Logical link control (LLC)
• used for storing the data in the buffers until
  media is accessed and data sent is acknowledged.
• a. Simple stop-wait protocol
• assumes an error free channel with finite buffer
  capacity
• processing speed at the nodes
• prevents a sender from flooding the receiver.
• Senders send one frame at a time and then waits
  for ACK (acknowledgment) before proceeding to
  transfer next data.
• b. Stop-wait protocol with timers and ARQ
  (Automatic Repeat request)
• used in case of error prone channels
• Sender could send a frame and timeout, and send a
  frame again if it does not receive ACK from the
  receiver.

Cont.
29
LLC Protocols Cont.

• c. Sliding window protocols
• It considers two channels (forward and reverse)
  in which regular frames will be sent in forward
  channel and ACKs will be sent in the reverse
  channel. The protocol maintains a sender and
  receiver window.
• d. Go-back protocol
• sender sends N frames and waits for an
  acknowledgment, if rth frame is in error, it
  starts sending from rth to Nth frame and early
  packets received after rth frame will be discarded

30
Standard Data link Protocols

• HDLC (High level Data Link Control)
• packetization standard for serial links
  connecting the remote devices in the network with
  central computer either point-to-point or
  point-to-multipoint.
• supports sliding window mode for reliable
  delivery mode of operation among others
• SLIP (Serial Line IP)
• used to connect a computer to Internet over
  dial-up line using the modem.
• does not support error checking, correction,
  detection and authentication.
• It supports only IP networks and uses character
  stuffing for data framing.
• PPP (Point-to-Point Protocol)
• It is approved and widely used protocol to
  connect home PCs to Internet over dial-up lines.
• The protocol handles error detection.
• It has two components
• LCP (link control protocol) It is used for
  authentication, bringing up lines, negotiating
  and bring down lines when needed.
• NCP (network control protocol). This of handles
  negotiation with network layer and gets the IP
  address allocated at connection time. It supports
  multiple protocols.

31
Internet Protocol (IP)

• It is mainly concerned with addressing of network
  nodes, security, network quality of service,
  fragmentation of packets.
• An IP packet consists of sender and destination
  addresses to facilitate forwarding of packets.
• An IP address is 32 bits long, which is grouped
  eight bits at a time, separated by dots and
  represented in decimal format (dotted decimal
  notation).

32
IP Address Subnetting

• IP networks can be divided into smaller networks
  called subnetworks (or subnets).
• Subnetting provides the network administrator
  extra flexibility and efficient use of network
  addresses.

33
Example IP Address subnetting
34
IP Packet format
35
Routing

• Internet has been divided into logical clusters
  called as Autonomous Systems (ASs).
• An AS consists of group of networks and routers.
• Three types of ASes
• Stub AS Used for small corporation which has
  single connection to other ASes.
• Multihomed AS Used for large corporation (no
  transit) which has multiple connections to other
  ASes.
• Transit AS Used by service providers to hook
  many ASes together.

36
Internet Hierarchical Routing

• Intra-AS routing - Routing within the ASs
• Inter-AS routing - Routing between the ASs
• Internet routing protocols are classified into
  two types
• Intra-AS routing protocol The routing protocol
  run by a router in an AS to find the routes
  within AS is called as intra-AS routing protocol
• Inter-AS routing protocol routing protocol used
  by the gateway routers to find the paths between
  the ASs is called inter-AS routing protocol.
• Gateway routers
• of an AS are connected to routers of others ASs
  that run intra-AS routing protocol with all other
  routers in AS.
• The gateway router also runs intra-AS routing
  protocol to find routes within the AS.

37
Intra-AS Inter-AS Routing

• Three ASes, A, B, and C.
• AS A has four routers, A.a, A.b, A.c, and A.d,
  which run the intra-AS routing protocol used
  within autonomous system A, ASes B and C have
  three and two routers, respectively.
• The gateway routers are A.a, A.c, B.a, and C.b.
• In addition to running the intra-AS routing
  protocol in conjunction with other routers in
  their ASs, these four routers run an inter-AS
  routing protocol among themselves.

38
Example Routing Scenario
39
Example Routing Scenario...

• Host h1 attached to router A.d needs to route a
  packet to destination h2 in autonomous system B.
• The packet is first routed on the link connected
  to A.d to A.c using A's intra-AS routing
  protocol.
• Router A.c will receive the packet and see that
  it is destined to an autonomous system outside of
  A.
• A.c's routing table for the inter-AS protocol
  would indicate that a packet destined to
  autonomous system B should be routed along the
  A.c to B.a link.
• When the packet arrives at B.a, B.a's inter-AS
  routing sees that the packet is destined for
  autonomous system B.
• The packet is then "handed over" to the intra-AS
  routing protocol within B, which routes the
  packet to its final destination, h2.
• In Figure 2.18, the portion of the path routed
  using A's intra-AS protocol is shown on the lower
  plane with a dotted line, the portion using the
  inter-AS routing protocol is shown in the upper
  plane as a solid line, and the portion of the
  path routed using B's intra-AS protocol is shown
  on the lower plane with a dotted line.

40
Routing Protocols

• RIP (Routing Information Protocol)
• an interior gateway routing protocol
• computes the routes within an autonomous system
  (intra-AS routing)
• Each router maintains a routing table.
• This table contains one entry per destination
  network, representing the current best route to
  the destination.
• OSPF (Open Shortest Path First)
• interior gateway protocol
• For use internal to a single Autonomous System.
• OSPF uses link-state or SPF (Shortest Path
  First)-based technology
• Border Gateway Protocol (BGP)
• an inter-AS routing protocol
• for use between multiple autonomous systems.
• Hosts using BGP communicate using the TCP.

Cont..
41
Cont..

• Multicast protocols
• IP Multicasting uses class D addresses.
• Multicasting provides an efficient way of
  disseminating data from a sender to a group of
  receivers.
• Data destined for the receivers in a multicast
  group is sent to a single multicast address.

42
Cont..

• IGMP (Internet Group Management Protocol)
• Used by IP hosts to report their host group
  memberships to any immediately-neighboring
  multicast routers.
• IGMP is an asymmetric protocol
• BOOTP (Bootstrap Protocol)
• allows a diskless client machine to discover its
  own IP address, the address of a server host, and
  the name of a file to be loaded into memory and
  executed.
• DHCP (Dynamic Host Configuration Protocol)
• It is used to control vital networking parameters
  of hosts with the help of a server.
• DHCP is backward compatible with Bootstrap
  protocol.

43
Mobile IP

• Designed to solve problem of mobility.
• Allows mobile node to use two IP addresses
• home address is static and is used to identify
  TCP connections.
• care-of address changes at each new point of
  attachment and can be called as the mobile node's
  topologically significant addres
• Mobile IP is a way of performing three related
  functions
• Agent Discovery Mobility agents advertise their
  availability on each link for which they provide
  service.
• Registration When the mobile node is away from
  home, it registers its care-of address with its
  home agent.
• Tunneling In order for datagrams to be delivered
  to the mobile node when it is away from home, the
  home agent has to tunnel the datagrams to the
  care-of address.

44
Mobile IP Routing
45
Protocols

• ICMP (Internet Control Message Protocol)
• used for out-of-band messages related to network
  operation or mis-operation.
• Some of ICMP's functions are to
• Announce network errors.
• Assist Troubleshooting.
• CIDR (Classless Inter Domain Routing)
• more efficient allocation of IP addresses
• CIDR currently uses prefixes anywhere from 13 to
  27 bits.
• A CIDR address includes the 32-bit IP address and
  information on how many bits are used for the
  network prefix.
• 206.13.01.48/25 the "/25" indicates the first
  25 bits are used to identify the unique network
  and the remaining bits identify the specific
  host.
• Used to overcome two problems
• Running short of IP addresses.
• Running out of capacity in the global routing
  tables.

46
CIDR Example

• If 1000 addresses were needed, we could supernet
  4 Class C networks (A, B, C and D) together as
  shown.
• Number of hosts connected in the network A, B, C
  and D are 241 (192.60.128.0 to .240), 251
  (192.60.129.0 to .250), 255 (192.60.130.0 to
  .254), and 253 (192.60.131.0 to .252)
  respectively.
• Subnet mask for this supernet is 255.255.252.0
  (22 higher bits to determine the supernet
  address).
• network portion is 22 bits long, the host portion
  is 10 bits long.
• The network address in CIDR 192.60.128.0/22.

47
RSVP

• RSVP (Resource Reservation Protocol)
• used to negotiate with network nodes to reserve
  bandwidth to support QoS path setup.
• RSVP based reservation
• host A sends a PATH message to host B, and host B
  sends a reservation message along the path to
  reserve the resources.
• In case of unavailable resources, RSVP sends an
  reservation error message to host B.

48
ARP RARP

• ARP (Address Resolution Protocol)
• maps IP address of a node to its corresponding
  MAC address
• To determine the 48 bit MAC address of the
  destination
• host sends broadcast packets onto the network
  asking "who owns IP address..."
• the machine with that address responds with the
  its MAC address.
• RARP (Reverse Address Resolution Protocol)
• maps the 48 bits MAC address to IP address.
• used for thin clients with limited facilities.
• Newly booted machine sends its 48 bit MAC address
  asking for its IP address to RARP server in the
  network.
• RARP server, after seeing the request, looks up
  the Ethernet address (MAC address) in its
  configuration files and sends back the
  corresponding IP address.

49
(No Transcript)
50
TCP Cont...
a) Establishment of a connection oriented session
using 3 way handshake mechanism. b) The
disconnection takes place using three service
primitives disc_request, disc_response and
disc_confirm after the transfer is over
51
TCP Cont...
d) Congestion control
c)TCP uses window management for traffic flow
control. d) Congestion Window vs. Transmission
Number
52
TCP Cont...

• RTT Estimation Algorithms
• Exponential averaging
• A smoothed RTT (SRTT) is computed by using the
  old esti- mate and the observed estimate.
• equations used in estimating SRTT and RTO when an
  acknowledgement for (i 1)th packet is received
• SRTT(i 1) a x RTT(i 1) (1 a) x SRTT(i)
• RTO SRTT(i 1) x d
• Where SRTT(i1) new estimate of RTT
• RTT currently measured RTT
• SRTT(i) old estimate of RTT and d2.
• The value of a must be less than 1.
• The a term determines how quickly the SRTT
  adapts to changes in the most recently measured
  RTT. It is empirically shown that values within
  0.8 and 0.9 give better RTT estimates.

53
TCP Cont...

• RTT Estimation Algorithms
• Jacobson algorithm
• similar to exponential averaging method.
• uses variance of a connection's RTT and
  incorporates this with an estimate of the RTT to
  set the value of the RTO.
• The following equations are used to find SRTT
  when an acknowledgment is received for an (i
  1)th packet.
• SRTT(i 1) (1 a) x SRTT(i) a x RTT(i 1)
• SERR(i 1) RTT(i 1) SRTT(i)
• SDEV (i 1) (1-h) x SDEV (i) h x SERR(i 1)
• RTO(i 1) SRTT(i 1) ß x SDEV(i1)
• Where SDEV smoothed delay variance, RTT
  observed round trip time, and SERR error in
  smoothing.
• Typical values for a, ß and h are 0.125, 4 and
  0.25, respectively.

54
TCP Cont...

• RTT Estimation Algorithms
• Karn Algorithm
• measures RTT for only those packets that are not
  retransmitted.
• When an acknowledgment arrives for a datagram
  that has been sent more than once, any RTT
  measurement based on this datagram is ignored.
• The algorithm uses more aggressive RTO
  (re-transmission timeout) backoff to enable the
  collection of accurate RTT measurements
  uncontaminated by retransmission ambiguity.
• The backed-off RTO for the retransmitted
  datagram is kept for the next datagram.
• The RTO is recalculated only when an
  acknowledgment arrives for a datagram that has
  not been retransmitted.

55
Transport Protocols

• UDP (User Datagram Protocol)
• connectionless transport of data.
• provides a way for applications to send
  encapsulated raw IP datagrams and without
  establishing a connection.
• does not provide reliability, flow control nor
  error recovery functions to IP.
• UDP headers contain fewer bytes (8 bytes) and
  consumes less network overhead than TCP.
• The UDP packet comprises of source port,
  destination port, length, the checksum and the
  data.
• Useful for real time multimedia applications by
  having error/flow control at the application
  layer.
• The protocol supports multicast of multimedia
  data.

56
Transport Protocols

• RTP (Real Time Protocol)
• supports multimedia data transfer over Internet
• uses UDP as a transport mechanism
• services include payload type identification
  (type of coding used), sequence num- bering,
  timestamping (generation time of packet) and
  delivery monitoring.
• supports data transfer to multiple destinations
  using multicast distribution.
• packet sequence numbers included in RTP allow the
  receiver to reconstruct the sender's packet
  sequence, but sequence numbers might also be used
  to determine the proper location of a packet,
  (for example in video decoding packet need not be
  in sequence.)
• It uses RTP control protocol (RTCP), to monitor
  the quality of service and to convey information
  about the participants in an on-going session.

57
Application Protocols

• Client-Server Model
• A client requests service from server.
• Server provides requested service.
• Eg. Web Browser as a client Web Server

Cont..
58
Application Protocols

• FTP (File Transfer Protocol)
• used to make reliable file transfers between the
  machines connected across the network by using
  TCP as a transport protocol.
• TFTP (Trivial File Transfer Protocol)
• used to perform unreliable file transfers between
  two machines connected across the network by
  using UDP as a transport protocol.

Cont..
59
Application Protocols

• E-mail
• Facilitates people to communicate with each other
  through the network.
• Three major components of E-mail are
• User agents for composing, editing, and reading
  mail messages.
• Mail servers to store outgoing, and incoming
  messages
• Simple mail transfer protocol (SMTP) a protocol
  between mail servers to exchange email messages,
  client is a sending mail server and server is the
  receiving mail server

60
Application Protocols

• Telnet
• used in emulating the terminal.
• facilitates a user to login to a remote machine
  connected across the network and emulate the
  remote terminal at the user machine.
• POP (Post Office Protocol)
• used to fetch e-mails from the remote mailbox and
  store it on the user's local machine to be read
  later.
• IMAP (Interactive Mail Access Protocol)
• A more sophisticated mail delivery protocol.
• It is designed to help the user who uses multiple
  computers, perhaps, a workstation in office, a PC
  at home, and a laptop on the road.
• maintains a central repository that can be
  accessed from any machine.

Cont..
61
Application Protocols

• SNMP (Simple Network Management Protocol)
• provides a systematic way of monitoring and
  managing a computer network.
• request-reply protocol running over UDP.

62
Application Protocols

• DNS (Domain Name System)
• Used for mapping host names and e-mail
  destinations to IP addresses.
• uses UDP connection to DNS server.
• three types of DNS servers
• Local name server,
• root name server
• Authoritative name server.

63
Application Protocols

• HTTP (Hyper Text Transfer Protocol)
• standard web transfer protocol used to retrieve
  the HTML documents stored across the hosts in the
  Internet.
• uses TCP for transport connection.