Transport Layer Comprehensive Notes textbook

Process-to-Process Communication: Understand the difference between process-to-process and host-to-host communication.
Addressing: Discuss the addressing mechanism at the transport layer, focusing on port numbers and their purposes.
Packetizing: Explain encapsulation and decapsulation of messages.
Multiplexing & Demultiplexing: Discuss services like multiplexing (many-to-one) and demultiplexing (one-to-many).
Flow Control: Explore how flow control can be managed at the transport layer.
Error Control: Discuss methods for error control at the transport layer.
Congestion Control: Understand how congestion control is implemented.
Connection Services: Differentiate between connectionless and connection-oriented services using Finite State Machines (FSMs).
Transport Protocols: Describe behaviors of four generic transport-layer protocols: simple, Stop-and-Wait, Go-Back-N, and Selective-Repeat protocols.
Bidirectional Communication: Understand the concept of bidirectional communication through piggybacking.

Layer Position: The transport layer operates between the network layer and the application layer, providing services to applications and utilizing services from the network layer.
Services Offered: Includes addressing, flow control, error control, and congestion control.

Transport Layers: Implementation ensures processes on hosts can communicate effectively.
Figures: Figure of network vs. transport layer shows how processes are managed.

Types of Port Numbers:
- Well-Known Ports: 0 through 1,023, assigned to specific protocols/services.
- Registered Ports: 1,024 to 49,151, registered with IANA for specific applications.
- Dynamic/Private Ports: 49,152 to 65,535 for non-registered applications.
Usage in UNIX: Well-known ports stored in /etc/services file, can be queried for application ports (e.g., TFTP on port 69).

Encapsulation Process: The process of adding headers to data as it moves down the protocol stack.
Decapsulation Process: The reverse process as data ascends through the protocol stack.

Multiplexing: The process where multiple applications can send their data over a single transport channel.
Demultiplexing: On the receiving side, separating the data back to the original applications.

Concept: Controls the rate of data transmission between sender and receiver.
Techniques:
- Single-slot buffering was discussed for its inefficiency.
- Concepts of pushing vs. pulling data during transmission.

Mechanisms: Use of sequence numbers for packets and acknowledgment to ensure data integrity.
Protocol Examples: Different methods can be employed based on the protocol being used (e.g., Stop-and-Wait, Go-Back-N, Selective-Repeat).

Definition: Techniques used to control network traffic that can lead to congestion.

Connection-Oriented Services: Establishes a connection before transmitting data. Includes extensive acknowledgment systems.
Connectionless Services: Data is sent without establishing a connection, with simple packet delivery techniques.

Transport Protocol Types:
- Simple Protocol: No flow or error control; meant for basic communication.
- Stop-and-Wait Protocol: Uses acknowledgment for received packets; packets can be resent on errors.
- Go-Back-N Protocol: Handles multiple packets but requires retransmission of all outstanding packets when an error occurs.
- Selective-Repeat Protocol: Resends only lost packets based on arrived acknowledgments, efficient in error recovery.

Concept: Allows data to be sent back and forth using the same channel efficiently by attaching acknowledgments to the data packets being sent back.

Numerous figures illustrate flow control, error control, encapsulation, and the functioning of different transport-layer protocols.

Understanding the transport layer is crucial for efficient data communication and management across networks. Different protocols serve varied networking needs, incorporating methods for control, reliability, and effective data packet management.