UDP Features

 

UDP Features

UDP has several important features that distinguish it from TCP. The major features discussed  are connectionless service, lack of error control, and lack of congestion control. Each can be either an advantage or a disadvantage depending on the application.

1. Connectionless Service

UDP is a connectionless protocol. Each UDP packet is independent of other packets sent by the same application.

There is:

  • No connection establishment
  • No connection termination
  • No relationship between successive packets
  • No sequencing of packets

Advantage

Connectionless communication is useful when the application needs to send a short request and receive a short response.

For example, in DNS, a client generally sends a short request and receives a short response. Both can usually fit into one UDP datagram.

Only two packets are required:

Client                         Server

Request  ───────────────────→
Response ←────────────────────

There is no need to spend time establishing and terminating a connection.

In contrast, a connection-oriented communication involves additional packets for connection establishment and termination. Therefore, UDP can provide lower delay for short request-response applications.

Example: DNS

DNS is a good example of an application that benefits from UDP.

  • Client sends a short query.
  • Server sends a short response.
  • The exchange can be completed using one datagram in each direction.
  • There is little concern about the ordering of multiple datagrams.

Therefore, the connectionless nature of UDP is an advantage for DNS.

Disadvantage

Connectionless service becomes a disadvantage when an application needs to send a long message.

For example, consider SMTP.

An e-mail may contain:

  • large amounts of text,
  • images,
  • audio,
  • video, etc.

If the message cannot fit into one UDP datagram, the application has to divide it into multiple datagrams.

Since UDP does not number or relate these datagrams, they may arrive at the receiver out of order.

Sender                         Receiver

Datagram 1 ─────────────────→  1
Datagram 2 ─────────────────→  3
Datagram 3 ─────────────────→  2

The receiving application may not be able to correctly reconstruct the original message.

Therefore, UDP's connectionless feature is not suitable for applications transmitting long messages that require reliable, ordered delivery.

2. Lack of Error Control

UDP does not provide error control. It provides an unreliable service, with the checksum providing error detection.

If a packet is lost or corrupted, UDP does not retransmit it.

Why can this be an advantage?

Reliable protocols such as TCP retransmit lost or corrupted packets.

Although retransmission improves reliability, it can introduce uneven delays.

Consider a real-time application transmitting:

Frame 1 → Frame 2 → Frame 3 → Frame 4

If Frame 2 is lost, a reliable protocol may retransmit Frame 2.

The application may have to wait for Frame 2 before continuing, producing a delay.

For some real-time applications, this delay is worse than simply losing one frame.

Example: Large Text File

Suppose we download a very large text file.

Part 1 → Part 2 → Part 3 → Part 4 → ...

If Part 2 is lost, we definitely want it to be retransmitted.

We are willing to wait for the missing part because the final file must be complete and correct.

Therefore:

UDP is not suitable for this type of application.

A reliable transport protocol such as TCP is more appropriate.

Example: Real-Time Interactive Application

Consider a real-time interactive application such as Skype, 

Audio and video are divided into frames:

Frame 1 → Frame 2 → Frame 3 → Frame 4 → ...

Suppose Frame 2 is corrupted.

If the transport protocol retransmits Frame 2, it may arrive after Frame 3 or Frame 4.

This can disturb the synchronization of the real-time transmission.

With UDP, the corrupted or lost datagram can simply be ignored, allowing the remaining frames to continue to the application.

The user may see a small temporary blank area or missing portion, but the overall transmission continues without waiting for retransmission.

Thus, for some real-time interactive applications, the lack of error control in UDP can actually be an advantage.

3. Lack of Congestion Control

UDP does not provide congestion control.

It does not regulate the transmission rate based on the level of congestion in the network.

Advantage

The lack of retransmission can sometimes prevent UDP from making congestion worse.

Consider a congested network:

Network Congestion
        ↓
Packet Loss
        ↓
TCP retransmits
        ↓
More packets enter the network
        ↓
Congestion may increase

TCP may retransmit lost packets several times. These additional packets can contribute to congestion.

UDP does not automatically retransmit lost packets.

Therefore, in some situations, the lack of error control and congestion-control mechanisms in UDP can be considered an advantage when congestion is a major concern.

UDP Features: Summary

FeatureAdvantageDisadvantage
Connectionless serviceLow delay; ideal for short request-response communicationLong messages may arrive out of order
Lack of error controlNo retransmission delay; useful for real-time applicationsLost/corrupted data is not recovered
Lack of congestion controlDoes not generate retransmissions that can worsen congestionUDP itself does not regulate network load

Important Examples 

ApplicationUDP SuitabilityReason
DNSSuitableShort request and quick response
SMTPNot suitableMessages can be long and require reliable delivery
Large file downloadNot suitableMissing/corrupted data must be recovered
Real-time audio/videoSuitable in many casesAvoids retransmission delays
Interactive applications such as SkypeSuitableTimely delivery is more important than perfect reliability

Key Point

UDP sacrifices reliability and control mechanisms in exchange for simplicity, low overhead, and low delay.

Therefore, whether a UDP feature is an advantage or disadvantage depends on the requirements of the application.

Comments

Post a Comment

Popular posts from this blog

Computer Networks PCCST501 Semester 5 KTU CS 2024 Scheme - Dr Binu V P

About Me  - Dr Binu  V P Model Question Paper Computer Networks PCCST501 KTU 2024 Scheme Syllabus Computer Networks PCCST501 KTU 2024 Scheme Do the programs to better understand the concept. Networks Lab PCCSL507 Module I Overview of the Computer Network and Internet Introduction to Computer Networks LAN ( Local Area Network ) WAN ( Wide Area Network) Comparison LAN and WAN Introduction to Internetwork Switched Network ( Circuit Switched and Packet Switched Network ) Comparison between Circuit Switching and Packet Switching Introduction to the Internet Brief History of the Internet Protocol Layering Introdcution to Protocol Layering TCP/IP Protocol Suite Functions of Each Layer in the TCP/IP Protocol Suite Encapsulation and Decapsulation in Protocol Layering Addressing in Protocol Layering Multiplexing and Demultiplexing in Protocol Layering ISO and the OSI Model Detailed Comparison of TCP/IP and OSI Model Application Layer Introduction to the Application Layer and ...
Read more

Introduction to Computer Networks

  Introduction to Computer Networks What is a Computer Network? A computer network is a collection of interconnected devices that can communicate and share data with each other. These devices may include: Desktop computers Laptops Smartphones Servers Printers Routers Switches The connection between devices can be established using: Wired media → cables such as Ethernet or fiber optics Wireless media → Wi-Fi, Bluetooth, cellular communication, etc. Simple Definition A computer network is the interconnection of devices capable of exchanging information. Why Do We Need Computer Networks? Computer networks are used to: Share Resources Printers Files Internet connection Enable Communication Email Video conferencing Messaging Share Data Databases Cloud storage Multimedia content Improve Reliability and Efficiency Backup systems Distributed processing Example of a Small Network When two computers at home a...
Read more

Introduction to Local Area Network (LAN)

Image
  Introduction to Local Area Network (LAN) What is a LAN? A Local Area Network (LAN) is a network that connects devices within a limited geographical area such as: A home An office A building A school campus A LAN is usually privately owned and managed by an individual or organization. Definition A LAN is a network that interconnects hosts in a single office, building, or campus to enable communication and resource sharing. Characteristics of a LAN 1. Limited Geographical Area A LAN covers a small area compared to larger networks like WANs. Examples: Computer lab in a college Office network in a company Home Wi-Fi network 2. High-Speed Communication Since devices are located close to each other, LANs provide: Faster data transfer Lower delay Efficient communication 3. Private Ownership LANs are generally owned and maintained by: Schools Companies Universities Home users 4. Resource Sharing LANs allow multiple users to shar...
Read more