Ch 3.

Network architecture
and hardware

Bcs sy
Contents
• Protocol stack design issues of the layers -addressing,
• error control,
• flow control,
• multiplexing and demultiplexing
• Routing
• Network Devices - NIC Cards, Hub, Switch, Bridges, Gateways,
Repeaters Router.
Error control
• Error control in data link layer is the process of detecting and
correcting data frames that have been corrupted or lost during
transmission.
• In case of lost or corrupted frames, the receiver does not
receive the correct data-frame and sender is ignorant about the
loss.
• Data link layer follows a technique to detect transit errors and
take necessary actions, which is retransmission of frames
whenever error is detected or frame is lost. The process is
called Automatic Repeat Request (ARQ).
Phases in Error Control
• The error control mechanism in data link layer involves the
following phases −
• Detection of Error − Transmission error, if any, is detected by
either the sender or the receiver.
• Acknowledgment − acknowledgment may be positive or
negative.
• Positive ACK − On receiving a correct frame, the receiver sends a
positive acknowledge.
• Negative ACK − On receiving a damaged frame or a duplicate frame,
the receiver sends a negative acknowledgment back to the sender.
Retransmission − The sender maintains a clock and sets a
timeout period. If an acknowledgment of a data-frame previously
transmitted does not arrive before the timeout, or a negative
acknowledgment is received, the sender retransmits the frame.
Error Control Techniques
There are three main techniques for error control −
Stop and Wait ARQ
• This protocol involves the following transitions −
• A timeout counter is maintained by the sender, which is started
when a frame is sent.
• If the sender receives acknowledgment of the sent frame within
time, the sender is confirmed about successful delivery of the
frame. It then transmits the next frame in queue.
• If the sender does not receive the acknowledgment within time,
the sender assumes that either the frame or its
acknowledgment is lost in transit. It then retransmits the frame.
• If the sender receives a negative acknowledgment, the sender
retransmits the frame
Go-Back-N ARQ
• The working principle of this protocol is −
• The sender has buffers called sending window.
• The sender sends multiple frames based upon the sending-window
size, without receiving the acknowledgment of the previous ones.
• The receiver receives frames one by one. It keeps track of incoming
frame’s sequence number and sends the corresponding
acknowledgment frames.
• After the sender has sent all the frames in window, it checks up to
what sequence number it has received positive acknowledgment.
• If the sender has received positive acknowledgment for all the
frames, it sends next set of frames.
• If sender receives NACK or has not receive any ACK for a particular
frame, it retransmits all the frames after which it does not receive any
positive ACK.
Selective Repeat ARQ
• Both the sender and the receiver have buffers called sending
window and receiving window respectively.
• The sender sends multiple frames based upon the sending-window
size, without receiving the acknowledgment of the previous ones.
• The receiver also receives multiple frames within the receiving
window size.
• The receiver keeps track of incoming frame’s sequence numbers,
buffers the frames in memory.
• It sends ACK for all successfully received frames and sends NACK
for only frames which are missing or damaged.
• The sender in this case, sends only packet for which NACK is
received.
Flow control
• It is technique that generally observes proper flow of data from
sender to receiver.
• It is very essential because it is possible for sender to transmit data
or information at very fast rate and hence receiver can receive this
information and process it.
• This can happen only if receiver has very high load of traffic as
compared to sender, or if receiver has power of processing less as
compared to sender.
• Flow control is basically technique that gives permission to two of
stations that are working and processing at different speeds to just
communicate with one another.
• Flow control in Data Link Layer simply restricts and coordinates
number of frames or amount of data sender can send just before it
waits for an acknowledgment from receiver.
• The receiving device also contains only limited amount of speed and
memory to store data.
• Approaches to Flow Control :
Flow Control is classified into two categories –
Feedback – based Flow Control
• In this control technique, sender simply transmits data or information
or frame to receiver, then receiver transmits data back to sender and
also allows sender to transmit more amount of data or tell sender
about how receiver is processing or doing. This simply means that
sender transmits data or frames after it has received
acknowledgments from user
Rate – based Flow Control
• In this control technique, usually when sender sends or transfer data
at faster speed to receiver and receiver is not being able to receive
data at the speed, then mechanism known as built-in mechanism in
protocol will just limit or restricts overall rate at which data or
information is being transferred or transmitted by sender without any
feedback or acknowledgment from receiver.
Techniques of Flow Control
• There are basically two types of techniques being developed to
control the flow of data –
Stop-and-Wait Flow Control
• This method is the easiest and simplest form of flow control. In this
method, basically message or data is broken down into various
multiple frames, and then receiver indicates its readiness to receive
frame of data. When acknowledgment is received, then only sender
will send or transfer the next frame.
Sliding Window Flow Control
• This method is required where reliable in-order delivery of packets or
frames is very much needed like in data link layer. It is point to point
protocol that assumes that none of the other entity tries to
communicate until current data or frame transfer gets completed. In
this method, sender transmits or sends various frames or packets
before receiving any acknowledgment.
multiplexing and demultiplexing
• Multiplexing –
Gathering data from multiple application processes of sender,
enveloping that data with header and sending them as a whole to the
intended receiver is called as multiplexing.
• Demultiplexing –
Delivering received segments at receiver side to the correct app layer
processes is called as demultiplexing.
• demultiplexing.
• Figure – Abstract view of multiplexing and demultiplexing
•
• Multiplexing and demultiplexing are the services facilitated by the
transport layer of OSI model.
• Figure – Transport layer- junction for multiplexing and demultiplexing
• There are two types of multiplexing and Demultiplexing :
• Connectionless Multiplexing and Demultiplexing
• Connection-Oriented Multiplexing and Demultiplexing
For sending data from an application at sender side to an application at
the destination side, sender must know the IP address of destination
and port number of the application (at the destination side) to which
he want to transfer the data
• Block diagram is shown below :
• Figure – Transfer of packet between applications of sender and
receiver
• Let us consider two messaging apps that are widely used now a days
viz. Hike and whatsapp.
• Suppose A is the sender and B is the receiver. Both sender and
receiver have these applications installed in their system (say
smartphone).
• Suppose A want to send messages to B in whatsapp and hike both. In
order to do so, A must mention the IP address of B and destination
port number of the whatsapp while sending the message through
whatsapp application.
• Similarly, for the later case, A must mention the IP address of B and
destination port number of the hike while sending the message
Routing
• A Router is a process of selecting path along which the data can be
transferred from source to the destination. Routing is performed by a
special device known as a router.
• A Router works at the network layer in the OSI model and internet
layer in TCP/IP model
• A router is a networking device that forwards the packet based on the
information available in the packet header and forwarding table.
• The routing algorithms are used for routing the packets. The routing
algorithm is nothing but a software responsible for deciding the optimal
path through which packet can be transmitted.
• The routing protocols use the metric to determine the best path for the
packet delivery. The metric is the standard of measurement such as
hop count, bandwidth, delay, current load on the path, etc. used by the
routing algorithm to determine the optimal path to the destination.
• The routing algorithm initializes and maintains the routing table for the
process of path determination.
Types of Routing
• Routing can be classified into three categories:
• Static Routing
• Default Routing
• Dynamic Routing
• Static Routing
• Static Routing is also known as Nonadaptive Routing.
• It is a technique in which the administrator manually adds the routes in
a routing table.
• A Router can send the packets for the destination along the route
defined by the administrator.
• In this technique, routing decisions are not made based on the
condition or topology of the networks
• Advantages Of Static Routing
• Following are the advantages of Static Routing:
• No Overhead: It has ho overhead on the CPU usage of the router.
Therefore, the cheaper router can be used to obtain static routing.
• Bandwidth: It has not bandwidth usage between the routers.
• Security: It provides security as the system administrator is allowed
only to have control over the routing to a particular network.
• Disadvantages of Static Routing:
• Following are the disadvantages of Static Routing:
• For a large network, it becomes a very difficult task to
add each route manually to the routing table.
• The system administrator should have a good
knowledge of a topology as he has to add each route
manually.
• Default Routing
• Default Routing is a technique in which a router is
configured to send all the packets to the same hop
device, and it doesn't matter whether it belongs to a
particular network or not. A Packet is transmitted to the
device for which it is configured in default routing.
• Default Routing is used when networks deal with the
single exit point.
• It is also useful when the bulk of transmission networks
have to transmit the data to the same hp device.
• When a specific route is mentioned in the routing table,
the router will choose the specific route rather than the
default route. The default route is chosen only when a
specific route is not mentioned in the routing table.
• Dynamic Routing
• It is also known as Adaptive Routing.
• It is a technique in which a router adds a new route in
the routing table for each packet in response to the
changes in the condition or topology of the network.
• Dynamic protocols are used to discover the new routes
to reach the destination.
• In Dynamic Routing, RIP and OSPF are the protocols
used to discover the new routes.
• If any route goes down, then the automatic adjustment
will be made to reach the destination.
• Advantages of Dynamic Routing:
• It is easier to configure.
• It is more effective in selecting the best route in
response to the changes in the condition or topology.
• Disadvantages of Dynamic Routing:
• It is more expensive in terms of CPU and bandwidth
usage.
• It is less secure as compared to default and static
routing.
Network device
HUBS
• Hubs are networking devices operating at a physical layer of the
OSI model that are used to connect multiple devices in a
network. They are generally used to connect computers in a
LAN.
• A hub has many ports in it. A computer which intends to be
connected to the network is plugged in to one of these ports.
When a data frame arrives at a port, it is broadcast to every
other port, without considering whether it is destined for a
particular destination device or not.
not.
• Features of Hubs
• A hub operates in the physical layer of the OSI model.
• A hub cannot filter data. It is a non-intelligent network device that
sends message to all ports.
• It primarily broadcasts messages. So, the collision domain of all
nodes connected through the hub stays one.
• Transmission mode is half duplex.
• Collisions may occurs during setup of transmission when more than
one computers place data simultaneously in the corresponding ports.
• Since they lack intelligence to compute best path for transmission of
data packets, inefficiencies and wastage occur.
• They are passive devices, they don’t have any software associated
with it.
• They generally have fewer ports of 4/12.
• Types of Hubs

Passive Hubs − Passive hubs connects nodes in a star

configuration by collecting wiring from nodes. They broadcast
signals onto the network without amplifying or regenerating them.
As they cannot extend the distance between nodes, they limit the
size of the LAN.
• Active Hubs − Active hubs amplify and regenerate the
incoming electrical signals before broadcasting them. They
have their own power supply and serves both as a repeater as
well as connecting centre. Due to their regenerating capabilities,
they can extend the maximum distance between nodes, thus
increasing the size of LAN.
• Intelligent Hubs − Intelligent hubs are active hubs that provide
additional network management facilities. They can perform a
variety of functions of more intelligent network devices like
network management, switching, providing flexible data rates
etc.
Switch