CS3591 UNIT 2

UNIT 2
TRANSPORT LAYER
1. INTRODUCTION
The transport layer is the fourth layer of the OSI model and is the core of
the Internet model. It responds to service requests from the session layer and
issues service requests to the network Layer. The transport layer provides
transparent transfer of data between hosts.
It provides end-to-end control and information transfer with the quality
o f service needed by the application program.
It is the first true end-to-end layer, implemented in all End Systems (ES).

2. TRANSPORT SERVICES
The transport layer is located between the network layer and the application
layer. The transport layer is responsible for providing services to the
application layer; it receives from the network layer. The services that can be
provided by the transport layer are
 Process-to-Process Communication
 Addressing: Port Numbers
 Encapsulation and Decapsulation
 Flow Control
 Error Control

1
CS3591 UNIT 2

 Congestion Control
 Multiplexing and demultiplexing
Process-to-Process Communication
The Transport Layer is responsible for delivering data to the appropriate
application process on the host computers. This involves multiplexing of data
from different application processes, i.e., forming data packets, and adding
source and destination port numbers in the header of each Transport Layer data
packet. Together with the source and destination IP address, the port numbers
constitute a network socket, i.e. an identification address of the process-to-
process communication.
Addressing: Port Numbers
Ports are the essential ways to address multiple entities in the same
location. Using port addressing it is possible to use more than one network-
based application at the same time.
Three types of Port numbers are used:
 Well-known ports - These are permanent port numbers. They range
between 0 to 1023.These port numbers are used by Server Process.
 Registered ports - The ports ranging from 1024 to 49,151 are not
assigned or controlled.
 Ephemeral ports (Dynamic Ports) – These are temporary port numbers.
They range between 49152–65535.These port numbers are used by Client
Process.
Encapsulation and Decapsulation
To send a message from one process to another, the transport-layer
protocol encapsulates and decapsulates messages.
Encapsulation happens at the sender site. The transport layer receives the data
and adds the transport-layer header. Decapsulation happens at the receiver site.
When the message arrives at the destination transport layer, the header is
dropped and the transport layer delivers the message to the process running at
the application layer.
Multiplexing and Demultiplexing
Whenever an entity accepts items from more than one source, this is
referred to as multiplexing (many to one).
Whenever an entity delivers items to more than one source, this is
referred to as demultiplexing (one to many). The transport layer at the source
performs multiplexing. The transport layer at the destination performs
demultiplexing
2
CS3591 UNIT 2

3. TRANSPORT LAYER PROTOCOLS

Three protocols are associated with the Transport layer. They are
1. UDP
2. TCP
3. SCTP
Each protocol provides a different type of service and should be used
appropriately.
UDP - UDP is an unreliable connectionless transport-layer protocol used
for its Simplicity and efficiency in applications where error control can be
provided by the application-layer process.
TCP - TCP is a reliable connection-oriented protocol that can be used in
any application where reliability is important.
SCTP - SCTP is a new transport-layer protocol designed to combine
some features of UDP and TCP in an effort to create a better protocol for
multimedia communication.

4. USER DATAGRAM PROTOTOCOL

User Datagram Protocol (UDP) is a connectionless, unreliable transport
protocol. UDP adds process-to-process communication to best-effort service
provided by IP. UDP is a very simple protocol using a minimum of overhead.
UDP is a simple demultiplexer, which allows multiple processes on each
host to communicate. UDP does not provide flow control, reliable or ordered
delivery. UDP can be used to send small message where reliability is not
expected. Sending a small message using UDP takes much less interaction
between the sender and receiver. UDP allow processes to indirectly identify
each other using an abstract locator called port or mailbox
UDP PORTS

3
CS3591 UNIT 2

 Processes (server/client) are identified by an abstract locator known

as port.
 Server accepts message at well known port.
 Some well-known UDP ports are 7–Echo, 53–DNS, 111–RPC,
161–SNMP, etc.
 < port, host > pair is used as key for demultiplexing.
 Ports are implemented as a message queue.
 When a message arrives, UDP appends it to end of the queue.
 When queue is full, the message is discarded.
 When a message is read, it is removed from the queue.
 When an application process wants to receive a message, one
is removed from the front of the queue.
If the queue is empty, the process blocks until a message becomes available

UDP DATA PACKET FORMAT

 UDP packets are known as user datagrams .
 These user datagrams, have a fixed-size header of 8 bytes made of
four fields, each of 2 bytes (16 bits).

4
CS3591 UNIT 2

Source port number

 Port number used by process on source host with 16 bits long.
 If the source host is client (sending request) then the port number is
a temporary one requested by the process and chosen by UDP
 If the source is server (sending response) then it is well known port
number.

Destination port number

 Port number used by process on Destination host with 16 bits long.
 If the destination host is the server (a client sending request) then the
port number is a well-known port number.
 If the destination host is client (a server sending response) then port
number is a temporary one copied by server from the request
packet.

Length
 This field denotes the total length of the UDP Packet (Header plus
data)
 The total length of any UDP datagram can be from 0 to 65,535
bytes.

Checksum
 UDP computes its checksum over the UDP header, the contents
of the message body, and something called the pseudo header.
 The pseudo header consists of three fields from the IP header—
protocol number, source IP address, destination IP address plus the
UDP length field
Data
 Data field defines that actual payload to be transmitted.
 Its size is variable.
UDP SERVICES
Process-to-Process Communication
UDP provides a process-to-process communication using socket addresses, a
combination of IP addresses and port numbers.
Connectionless Services
 UDP provides a connectionless service.
 There is no connection establishment and no connection
termination.
5
CS3591 UNIT 2

 Each user datagram sent by UDP is an independent datagram.

 There is no relationship between the different user datagrams even
if they are coming from the same source process and going to the
same destination program.
 The user datagrams are not numbered.
 Each user datagram can travel on a different path.
Flow Control
 UDP is a very simple protocol.
 There is no flow control, and hence no window mechanism.
 The receiver may overflow with incoming messages.
 The lack of flow control means that the process using UDP
should provide for this service, if needed.
Error Control
 There is no error control mechanism in UDP except for the
checksum.
 This means that the sender does not know if a message has been lost
or duplicated.
 When the receiver detects an error through the checksum, the
user datagram is silently discarded.
 The lack of error control means that the process using UDP
should provide for this service, if needed.

Checksum
 UDP checksum calculation includes three sections: a pseudo
header, the UDP header, and the data coming from the application
layer.
 The pseudo header is the part of the header in which the user
datagram is to be encapsulated with some fields filled with 0s.
Congestion Control
 Since UDP is a connectionless protocol, it does not provide
congestion control.
 UDP assumes that the packets sent are small and sporadic
(occasionally or at irregular intervals) and cannot create
congestion in the network.

6
CS3591 UNIT 2

 This assumption may or may not be true, when UDP is used for
interactive real-time transfer of audio and video.

Encapsulation and Decapsulation

 To send a message from one process to another, the UDP protocol
encapsulates and decapsulates messages.

Queuing
 In UDP, queues are associated with ports.
 At the client site, when a process starts, it requests a port number
from the operating system.
 Some implementations create both an incoming and an outgoing
queue associated with each process.
 Other implementations create only an incoming queue associated
with each process.

Multiplexing and Demultiplexing

 In a host running a transportprotocol suite, there is only one
UDP but possibly several processes that may want to use the
services of UDP.
 To handle this situation, UDP multiplexes and demultiplexes.
APPLICATIONS OF UDP
 UDP is used for management processes such as SNMP.
 UDP is used for route updating protocols such as RIP.
 UDP is a suitable transport protocol for multicasting. Multicasting
capability is embedded in the UDP software
 UDP is suitable for a process with internal flow and error control
mechanisms such as Trivial File Transfer Protocol (TFTP).
 UDP is suitable for a process that requires simple request-response
communication with little concern for flow and error
control.
 UDP is normally used for interactive real-time applications that
cannot tolerate uneven delay between sections of a received
message.

7
CS3591 UNIT 2

5. TRANSMISSION CONTROL PROTOCOL

TCP is a reliable, connection-oriented, byte-stream protocol’s
guarantees the reliable, in-order delivery of a stream of bytes. It is a full-duplex
protocol, meaning that each TCP connection supports a pair of byte streams,
one flowing in each direction.
TCP includes a flow-control mechanism for each of these byte
streams that allow the receiver to limit how much data the sender can transmit
at a given time. TCP supports a demultiplexing mechanism that allows multiple
application programs on any given host to simultaneously carry on a
conversation with their peers.
TCP also implements congestion-control mechanism. The idea of
this mechanism is to prevent sender from overloading the network. Flow
control is an end- t o - e n d issue, whereas congestion control is concerned
with how host and network interact.
TCP SERVICES
Process-to-Process Communication: TCP provides process-to-process
communication using port numbers.
Stream Delivery Service: TCP is a stream-oriented protocol. TCP allows the
sending process to deliver data as a stream of bytes and allows the receiving
process to obtain data as a stream of bytes. TCP creates an environment in
which the two processes seem to be connected by an imaginary “tube” that
carries their bytes across the Internet. The sending process produces (writes to)
the stream and the receiving process consumes (reads from) it.
Full-Duplex Communication:

TCP offers full-duplex service, where data can flow in both directions at the
same time. Each TCP endpoint then has its own sending and receiving
buffer, and segments move in both directions.
Multiplexing and Demultiplexing:
TCP performs multiplexing at the sender and demultiplexing at the
receiver.

8
CS3591 UNIT 2

Connection-Oriented Service:
TCP is a connection-oriented protocol. A connection needs to be
established for each pair of processes. When a process at site A wants to send to
and receive data from another process at site B, the following three phases
occur:
 The two TCP’s establish a logical connection between them.
 Data are exchanged in both directions.
 The connection is terminated.
Reliable Service: TCP is a reliable transport protocol. It uses an
acknowledgment mechanism to check the safe and sound arrival of data.
TCP SEGMENT
A packet in TCP is called a segment. Data unit exchanged between TCP
peers are called segments. A TCP segment encapsulates the data received from
the application layer. The TCP segment is encapsulated in an IP datagram,
which in turn is encapsulated in a frame at the data-link layer.

TCP is a byte-oriented protocol, which means that the sender writes

bytes into a TCP connection and the receiver reads bytes out of the TCP
connection. TCP does not, itself, transmit individual bytes over the Internet.
TCP on the source host buffers enough bytes from the sending process to
fill a reasonably sized packet and then sends this packet to its peer on the
destination host. TCP on the destination host then empties the contents of the
packet into a receive buffer, and the receiving process reads from this buffer at
its leisure. TCP connection supports byte streams flowing in both directions.
The packets exchanged between TCP peers are called segments, since each
one carries a segment of the byte stream.
TCP PACKET FORMAT
Each TCP segment contains the header plus the data. The segment
consists of a header of 20 to 60 bytes, followed by data from application
9
CS3591 UNIT 2

program. The header is 20 bytes if there are no options and up to 60 bytes if it

contains options.

TCP CONNECTION MANAGEMENT

TCP is connection-oriented. A connection-oriented transport protocol
establishes a logical path between the source and destination. All of the
segments belonging to a message are then sent over this logical path. In TCP,
connection-oriented transmission requires three phases: Connection
Establishment, Data Transfer and Connection Termination.

Connection Establishment
While opening a TCP connection the two nodes(client and server) want to
agree on a set of parameters.
The parameters are the starting sequence numbers that is to be used for their
respective byte streams. Connection establishment in TCP is a three-way
handshaking.

 Client sends a SYN segment to the server containing its initial sequence
number (Flags = SYN, Sequence Num = x)
 Server responds with a segment that acknowledges client’s
segment and specifies its initial sequence number (Flags = SYN +
ACK, ACK = x + 1 Sequence Num = y).
10
CS3591 UNIT 2

 Finally, client responds with a segment that acknowledges server’s

sequence number (Flags = ACK, ACK = y + 1).
 The reason that each side acknowledges a sequence number that is one
larger than the one sent is that the Acknowledgment field actually
identifies the “next sequence number expected,”
 A timer is scheduled for each of the first two segments, and if the
expected response is not received, the segment is retransmitted.
Data Transfer
 After connection is established, bidirectional data transfer can take place.
 The client and server can send data and acknowledgments in both
directions.
 The data traveling in the same direction as an acknowledgment are
carried on the same segment.
 The acknowledgment is piggy backed with the data.
Connection Termination
Connection termination or teardown can be done in two ways:
Three-way Close and Half-Close

Three-way Close—Both client and server close simultaneously.

Client sends a FIN segment. The FIN segment can include last chunk of data.
Server responds with FIN + ACK segment to inform its closing. Finally,
client sends an ACK segment

Half-Close—Client stops sending but receives data.

11
CS3591 UNIT 2

Client half-closes the connection by sending a FIN segment. Server sends an

ACK segment. Data transfer from client to the server stops. After sending all
data, server sends FIN segment to client, which is acknowledged by the client.

6. CONGESTION CONTROL
Congestion in a network may occur if the load on the network (the number of
packets sent to the network) is greater than the capacity of the network (the
number of packets a network can handle). Congestion control refers to the
mechanisms and techniques that control the congestion and keep the load below
the capacity. Congestion Control refers to techniques and mechanisms that
can either prevent congestion, before it happens, or remove congestion, after it
has happened Congestion control mechanisms are divided into two categories
1. Open loop
2. Closed loop
Congestion occurs if load (number of packets sent) is greater than capacity
of the network (number of packets a network can handle). When load is less
than network capacity, throughput increases proportionally. When load exceeds
capacity, queues become full and the routers discard some packets and
throughput declines sharply. When too many packets are contending for the
same link The queue overflows, Packets get dropped, Network is congested,
Network should provide a congestion control mechanism to deal with such a
situation.
Additive Increase / Multiplicative Decrease (AIMD)
 TCP source initializes Congestion Window based on congestion level in
the network. Source increases Congestion Window when level of
congestion goes down and decreases the same when level of congestion goes
up. TCP interprets timeouts as a sign of congestion and reduces the rate of
12
CS3591 UNIT 2

transmission. On timeout, source reduces its Congestion Window by half, i.e.,

multiplicative decrease. For example, if Congestion Window = 16 packets, after
timeout it is 8. Value of Congestion Window is never less than maximum
segment size (MSS). When ACK arrives Congestion Window is incremented
marginally, i.e., additive increase.
Increment = MSS × (MSS/Congestion Window) Congestion Window +=
Increment
For example, when ACK arrives for 1 packet, 2 packets are sent. When
ACK for both packets arrive, 3 packets are sent and so on. Congestion
Window increases and decreases throughout lifetime of the connection.

Slow Start
Slow start is used to increase Congestion Window exponentially from a cold
start. Source TCP initializes Congestion Window to one packet. TCP doubles
the number of packets sent every RTT on successful transmission. When ACK
arrives for first packet TCP adds 1 packet to Congestion Window and sends
two packets. When two ACKs arrive, TCP increments Congestion Window by
2 packets and sends four packets and so on. Instead of sending entire
permissible packets at once (burst traffic), packets are sent in a phased manner,
i.e., slow start. Initially TCP has no idea about congestion,
henceforth it increases Congestion Window rapidly until there is a
timeout. On timeout:
CongestionThreshold = CongestionWindow/ 2 CongestionWindow = 1
Slow start is repeated until Congestion Window reaches Congestion
Threshold and thereafter 1 packet per RTT. The congestion window trace
will look like

13
CS3591 UNIT 2

Fast Retransmit and Fast Recovery

TCP timeouts led to long periods of time during which the connection
went dead while waiting for a timer to expire. Fast retransmit is a heuristic
approach that triggers retransmission of a dropped packet sooner than the
regular timeout mechanism. It does not replace regular timeouts. When a packet
arrives out of order, receiving TCP resends the same acknowledgment
(duplicate ACK) it sent last time. When three duplicate ACK arrives at the
sender, it infers that corresponding packet may be lost due to congestion and
retransmits that packet. This is called fast retransmit before regular timeout.
When packet loss is detected using fast retransmit, the slow start phase is
replaced by additive increase, multiplicative decrease method. This is known as
fast recovery. Instead of setting Congestion Window to one packet, this method
uses the ACKs that are still in pipe to clock the sending of packets. Slow start is
only used at the beginning of a connection and after regular timeout. At other
times, it follows a pure AIMD pattern.

14
CS3591 UNIT 2

7. CONGESTION AVOIDANCE
Congestion avoidance mechanisms prevent congestion before it actually
occurs. These mechanisms predict when congestion is about to happen and then
to reduce the rate at which hosts send data just before packets start being
discarded. TCP creates loss of packets in order to determine bandwidth of the
connection. Routers help the end nodes by intimating when congestion is likely
to occur. Congestion-avoidance mechanisms are:
DEC bit - Destination Experiencing Congestion Bit
RED - Random Early Detection

Dec Bit - Destination Experiencing Congestion Bit

The first mechanism developed for use on the Digital Network
Architecture(DNA). The idea is to evenly split the responsibility for
congestion control between the routers and the end nodes. Each router
monitors the load it is experiencing and explicitly notifies the end nodes when
congestion is about to occur. This notification is implemented by setting a
binary congestion bit in the packets that flow through the router; hence the
name DECbit.

15
CS3591 UNIT 2

The destination host then copies this congestion bit into the ACK it sends
back to the source. The Source checks how many ACK has DEC bit set for
previous window packets. If less than 50% of ACK have DEC bit set, then
source increases its congestion window by 1 packet, Otherwise, decreases the
congestion window by 87.5%. Finally, the source adjusts its sending rate so as
to avoid congestion. Increase by 1, decrease by 0.875 rule was based on AIMD
for stabilization. A single congestion bit is added to the packet header. Using a
queue length of 1 as the trigger for setting the congestion bit. A router sets this
bit in a packet if its average queue length is greater than or equal to 1 at the

time the packet arrives.

Average queue length is measured over a time interval that includes the
last busy + last idle cycle + current busy cycle.
It calculates the average queue length by dividing the curve area with time
interval.
Red - Random Early Detection
The second mechanism of congestion avoidance is called as
Random Early Detection (RED). Each router is programmed to monitor its own
queue length, and when it detects that there is congestion, it notifies the source
to adjust its congestion window. RED differs from the DEC bit scheme by two
ways: In DECbit, explicit notification about congestion is sent to source,
whereas RED implicitly notifies the source by dropping a few packets.
DECbit may lead to tail drop policy, whereas RED drops packet based on
drop probability in a random manner. Drop each arriving packet with some drop
probability whenever the queue length exceeds some drop level. This idea is
called early random drop.
Computation of average queue length using RED
Avg Len = (1 − Weight) × Avg Len + Weight × Sample Len
where 0 < Weight < 1 and Sample Len – is the length
16
CS3591 UNIT 2

of the queue when a sample measurement is made. The queue length is

measured every time a new packet arrives at the gateway.

8.STREAM CONTROL TRANSMISSION PROTOCOL

Stream Control Transmission Protocol (SCTP) is a reliable, message-oriented
transport layer protocol. SCTP has mixed features of TCP and UDP. SCTP
maintains the message boundaries and detects the lost data, duplicate data as
well as out-of-order data. SCTP provides the Congestion control as well as
Flow control. SCTP is especially designed for internet applications as well as
multimedia communication.
SCTP SERVICES
Process-to-Process Communication
SCTP provides process-to-process communication.
Multiple Streams
SCTP allows multistream service in each connection, which is called
association in SCTP terminology.

17
CS3591 UNIT 2

If one of the streams is blocked, the other streams can still deliver their

data.
Multihoming
An SCTP association supports multihoming service. The sending and receiving
host can define multiple IP addresses in each end for an association. In this
fault-tolerant approach, when one path fails, another interface can be used for
data delivery without interruption.

Full-Duplex Communication: SCTP offers full-duplex service, where data can

flow in both directions at the same time. Each SCTP then has a sending and
receiving buffer and packets are sent in both directions.
Connection-Oriented Service:
SCTP is a connection-oriented protocol. In SCTP, a connection is
called an association. If a client wants to send and receive message from
server , the steps are :
Step1: The two SCTPs establish the connection with each other.
Step2: Once the connection is established, the data gets exchanged in
both the directions.
18
CS3591 UNIT 2

Step3: Finally, the association is terminated.

Reliable Service
SCTP is a reliable transport protocol. It uses an acknowledgment
mechanism to check the safe and sound arrival of data.
SCTP PACKET FORMAT

An SCTP packet has a mandatory general header and a set of blocks called
chunks.
General Header
Source port
Destination port
Verification tag
Checksum
Chunks
Types of Chunks
An SCTP association may send many packets, a packet may contain
several chunks, and chunks may belong to different streams. SCTP defines two

types of chunks - Control chunks and Data chunks. A control chunk controls and
19
CS3591 UNIT 2

maintains the association. A data chunk carries user data.

SCTP ASSOCIATION
SCTP is a connection-oriented protocol. A connection in SCTP is called an
association to emphasize multihoming. SCTP Associations consists of three
phases:
 Association Establishment
 Data Transfer
 Association Termination
Association Establishment
Association establishment in SCTP requires a four-way handshake. In this
procedure, a client process wants to establish an association with a server
process using SCTP as the transport-layer protocol. The SCTP server needs to
be prepared to receive any association (passive open).
Association establishment, however, is initiated by the client (active open).

Data Transfer
The whole purpose of an association is to transfer data between two
ends. After the association is established, bidirectional data transfer
can take place. The client and the server can both send data. SCTP
supports piggybacking. Types of SCTP data Transfer :
Multihoming Data Transfer
Multistream Delivery
Association Termination
In SCTP,either of the two parties involved in exchanging data
(client or server) can close the connection. SCTP does not allow a “half
closed” association. If one end closes the association, the other end must
20
CS3591 UNIT 2

stop sending new data. If any data are left over in the queue of the
recipient of the termination request, they are sent and the association is
closed. Association termination uses three packets.

SCTP FLOW CONTROL

Flow control in SCTP is similar to that in TCP. In SCTP, we need to
handle two units of data, the byte and the chunk. The values of rwnd and cwnd
are expressed in bytes; the values of TSN and acknowledgments are expressed
in chunks. Current SCTP implementations still use a byte-oriented window for
flow control.
Receiver Site:
The receiver has one buffer (queue) and three variables. The queue holds
the received data chunks that have not yet been read by the process. The first
variable holds the last TSN received, cum TSN. The second variable holds the
available buffer size; win size. The third variable holds the last accumulative
acknowledgment, last ACK. The following figure shows the queue and
variables at the receiver site.

21
CS3591 UNIT 2

When the site receives a data chunk, it stores it at the end of the buffer (queue)
and subtracts the size of the chunk from win Size. The TSN number of the
chunk is stored in the cum TSN variable. 2. When the process reads a chunk, it
removes it from the queue and adds the size of the removed chunk to winSize
(recycling).
3. When the receiver decides to send a SACK, it checks the value of lastAck; if
it is less than cumTSN, it sends a SACK with a cumulative TSN number equal
to the cumTSN. It also includes the value of win Size as the advertised window
size.
Sender Site:
The sender has one buffer (queue) and three variables: curTSN, rwnd, and in
Transit, as shown in the following figure. We assume each chunk is 100 bytes
long. The buffer holds the chunks produced by the process that either have been
sent or are ready to be sent. The first variable, curTSN, refers to the next chunk
to be sent. All chunks in the queue with a TSN less than this value have been
sent, but not acknowledged; they are outstanding. The second variable, rwnd,
holds the last value advertised by the receiver (in bytes). The third variable, in
Transit, holds the number of bytes in transit, bytes sent but not yet
acknowledged. The following is the procedure used by the sender.
1. A chunk pointed to by curTSN can be sent if the size of the data is less than
or equal to the quantity rwnd - inTransit. After sending the chunk, the value of
curTSN is incremented by 1 and now points to the next chunk to be sent. The
value of inTransit is incremented by the size of the data in the transmitted
chunk.
2. When a SACK is received, the chunks with a TSN less than or equal to the
cumulative TSN in the SACK are removed from the queue and discarded. The
sender does not have to worry about them anymore. The value of inTransit is
reduced by the total size of the discarded chunks. The value of rwnd is updated
with the value of the advertised window in the SACK.

22
CS3591 UNIT 2

9. QUALITY OF SERVICE
Quality of service (QoS) is the use of mechanisms or technologies that work on
a network to control traffic and ensure the performance of critical applications
with limited network capacity. It enables organizations to adjust their
overall network traffic by prioritizing specific high-performance applications.
QoS is typically applied to networks that carry traffic for resource-intensive
systems. Common services for which it is required include internet protocol
television (IPTV), online gaming, streaming media, videoconferencing, video
on demand (VOD), and Voice over IP (VoIP). Using QoS in networking,
organizations have the ability to optimize the performance of multiple
applications on their network and gain visibility into the bit rate, delay, jitter,
and packet rate of their network. This ensures they can engineer the traffic on
their network and change the way that packets are routed to the internet or other
networks to avoid transmission delay. This also ensures that the organization
achieves the expected service quality for applications and delivers expected user
experiences. As per the QoS meaning, the key goal is to enable networks and
organizations to prioritize traffic, which includes offering dedicated bandwidth,
controlled jitter, and lower latency. The technologies used to ensure this are
vital to enhancing the performance of business applications, wide-area networks
(WANs), and service provider networks. QoS networking technology works by
marking packets to identify service types, then configuring routers to create
separate virtual queues for each application, based on their priority. As a result,
bandwidth is reserved for critical applications or websites that have been
23
CS3591 UNIT 2

assigned priority access. QoS technologies provide capacity and handling

allocation to specific flows in network traffic. This enables the network
administrator to assign the order in which packets are handled and provide the
appropriate amount of bandwidth to each application or traffic flow.

Types of Network Traffic

Understanding how QoS network software works is reliant on defining the

various types of traffic that it measures. These are:

Bandwidth: The speed of a link. QoS can tell a router how to use bandwidth.
For example, assigning a certain amount of bandwidth to different queues for
different traffic types.

Delay: The time it takes for a packet to go from its source to its end destination.
This can often be affected by queuing delay, which occurs during times of
congestion and a packet waits in a queue before being transmitted. QoS enables
organizations to avoid this by creating a priority queue for certain types of
traffic.

Loss: The amount of data lost as a result of packet loss, which typically occurs
due to network congestion. QoS enables organizations to decide which packets
to drop in this event.

Jitter: The irregular speed of packets on a network as a result of congestion,

which can result in packets arriving late and out of sequence. This can cause
distortion or gaps in audio and video being delivered.

24
CS3591 UNIT 2

Token bucket algorithm

Token bucket algorithm is one of the techniques for congestion control
algorithms. When too many packets are present in the network it causes packet
delay and loss of packet which degrades the performance of the system. This
situation is called congestion.
The network layer and transport layer share the responsibility for handling
congestions. One of the most effective ways to control congestion is trying to
reduce the load that transport layer is placing on the network. To maintain this
network and transport layers have to work together.
The Token Bucket Algorithm is diagrammatically represented as follows –

The leaky bucket algorithm enforces output patterns at the average rate, no
matter how busy the traffic is. So, to deal with the more traffic, we need a
flexible algorithm so that the data is not lost. One such approach is the token
bucket algorithm.
Let us understand this algorithm step wise as given below −
25
CS3591 UNIT 2

 Step 1 − In regular intervals tokens are thrown into the bucket f.

 Step 2 − The bucket has a maximum capacity f.
 Step 3 − If the packet is ready, then a token is removed from the
bucket, and the packet is sent.
 Step 4 − Suppose, if there is no token in the bucket, the packet
cannot be sent.
Leaky bucket algorithm
Leaky Bucket Algorithm mainly controls the total amount and the rate of the
traffic sent to the network.
Step 1 − Let us imagine a bucket with a small hole at the bottom where the rate
at which water is poured into the bucket is not constant and can vary but it
leaks from the bucket at a constant rate.
Step 2 − So (up to water is present in the bucket), the rate at which the water
leaks does not depend on the rate at which the water is input to the bucket.
Step 3 − If the bucket is full, additional water that enters into the bucket that
spills over the sides and is lost.
Step 4 − Thus the same concept applied to packets in the network. Consider
that data is coming from the source at variable speeds. Suppose that a source
sends data at 10 Mbps for 4 seconds. Then there is no data for 3 seconds. The
source again transmits data at a rate of 8 Mbps for 2 seconds. Thus, in a time
span of 8 seconds, 68 Mb data has been transmitted.
That’s why if a leaky bucket algorithm is used, the data flow would be 8 Mbps
for 9 seconds. Thus, the constant flow is maintained.

Differentiated QoS

The differentiated services (DS), or DiffServ, approach provides a simpler and

more scalable QoS. DS minimizes the amount of storage needed in a router by
processing traffic flows in an aggregate manner, moving all the complex
procedures from the core to the edge of the network. A traffic conditioner is one
of the main features of a DiffServ node to protect the DiffServ domain. As
shown in Figure 6.5, the traffic conditioner includes four major components:
meter, marker, shaper, and dropper.

26
CS3591 UNIT 2

A meter measures the traffic to make sure that packets do not exceed their
traffic profiles. A marker marks or unmarks packets in order to keep track of
their situations in the DS node. A shaper delays any packet that is not compliant
with the traffic profile. Finally, a dropper discards any packet that violates its
traffic profile.

Advantages of QoS
The deployment of QoS is crucial for businesses that want to ensure the
availability of their business-critical applications. It is vital for delivering
differentiated bandwidth and ensuring data transmission takes place without
interrupting traffic flow or causing packet losses. Major advantages of
deploying QoS include:

1. Unlimited application prioritization: QoS guarantees that businesses’

most mission-critical applications will always have priority and the
necessary resources to achieve high performance.
2. Better resource management: QoS enables administrators to better
manage the organization’s internet resources. This also reduces costs and
the need for investments in link expansions.
3. Enhanced user experience: The end goal of QoS is to guarantee the high
performance of critical applications, which boils down to delivering
optimal user experience. Employees enjoy high performance on their
27
CS3591 UNIT 2

high-bandwidth applications, which enables them to be more effective

and get their job done more quickly.
4. Point-to-point traffic management: Managing a network is vital
however traffic is delivered, be it end to end, node to node, or point to
point. The latter enables organizations to deliver customer packets in
order from one point to the next over the internet without suffering any
packet loss.
5. Packet loss prevention: Packet loss can occur when packets of data are
dropped in transit between networks. This can often be caused by a
failure or inefficiency, network congestion, a faulty router, loose
connection, or poor signal. QoS avoids the potential of packet loss by
prioritizing bandwidth of high-performance applications.
6. Latency reduction: Latency is the time it takes for a network request to
go from the sender to the receiver and for the receiver to process it. This
is typically affected by routers taking longer to analyse information and
storage delays caused by intermediate switches and bridges. QoS enables
organizations to reduce latency, or speed up the process of a network
request, by prioritizing their critical application.

28