Flow Control and Go-Back-N ARQ

1. Definition of Flow Control (2 marks)

Flow control is a set of procedures that tells the sender how much data or how many frames it can transmit before it must wait for an acknowledgement from the receiver. It is a technique that allows two stations working and processing at different speeds to communicate with each other without data loss or corruption.

: "Flow control is basically a technique that gives permission to two stations that are working and processing at different speeds to just communicate with another. It is a set of procedures that explains how much data or frames it can transfer before data overwhelms the receiver."

Why it is needed:

• The sender may be faster than the receiver.
• Without flow control, the receiver's buffer can overflow, causing frames to be lost or corrupted.
• It ensures smooth, reliable, and orderly data transmission between sender and receiver.

2. Go-Back-N ARQ (5 marks)

2.1 Concept

Go-Back-N ARQ (Automatic Repeat reQuest) is a sliding window protocol used for error control and flow control at the data link layer. Key characteristics are:

• The sender can transmit multiple frames before receiving an acknowledgement, up to a maximum window size of 2^n - 1 frames (where n is the number of bits in the sequence number field).
• The receiver accepts frames only in order. If a frame is received out of order or is corrupted, it is discarded.
• If an error is detected in frame i, the receiver discards frame i and all subsequent frames already received. The sender must go back and retransmit frame i and all frames that followed it.
• The receiver sends a Negative Acknowledgement (NAK) for the erroneous frame, or the sender uses a timeout mechanism.

2.2 Window Size

2.3 Working Mechanism

Step-by-step process:

1. Sender transmits frames 0, 1, 2, 3, ... up to the window size without waiting for individual ACKs.
2. Receiver checks each incoming frame for errors.
3. If a frame is received correctly and in order, the receiver sends ACK (next expected frame number).
4. If a frame is corrupted or lost, the receiver sends a NAK or simply discards it.
5. The sender, upon receiving NAK or on timeout, retransmits the erroneous frame and all frames sent after it.

2.4 Example

Assume:

• Sequence number bits = 3, so sequence numbers = 0, 1, 2, 3, 4, 5, 6, 7
• Window size = 2^3 - 1 = 7 frames
• Frames 0 through 6 are to be sent

Sender                              Receiver
  |--- Frame 0 ------------------>  |  ACK 1
  |--- Frame 1 ------------------>  |  ACK 2
  |--- Frame 2 ------------------>  |  ACK 3
  |--- Frame 3 (LOST/ERROR) ---X    |
  |--- Frame 4 ------------------>  |  (Discarded - out of order)
  |--- Frame 5 ------------------>  |  (Discarded - out of order)
  |--- Frame 6 ------------------>  |  (Discarded - out of order)
  |<-- NAK 3 (or Timeout) --------|
  |--- Frame 3 (Retransmit) ----->  |  ACK 4
  |--- Frame 4 (Retransmit) ----->  |  ACK 5
  |--- Frame 5 (Retransmit) ----->  |  ACK 6
  |--- Frame 6 (Retransmit) ----->  |  ACK 7

Explanation of the example:

• Frames 0, 1, 2 are received correctly and acknowledged.
• Frame 3 is lost in transmission.
• Frames 4, 5, 6 arrive at the receiver but are discarded because the receiver is waiting for Frame 3 (receiver window size = 1, strictly in-order).
• The sender receives NAK 3 (or the timer expires), and it goes back to Frame 3 and retransmits Frames 3, 4, 5, and 6.

2.5 Efficiency of Go-Back-N

• If no error: Efficiency = W / (1 + 2a) where W = window size, a = propagation delay / transmission time
• If error occurs: Efficiency is reduced because multiple frames are retransmitted unnecessarily.

3. Difference Between Go-Back-N ARQ and Stop-and-Wait ARQ (3 marks)

Stop-and-Wait is in fact the special case of Go-Back-N with a sender window of one, which is why both use the same acknowledgement and timeout machinery and differ only in how many frames may be outstanding at once.

4. Conclusion

Flow control is the set of procedures that stops a fast sender from overwhelming a slow receiver by limiting how much unacknowledged data may be in transit, and when it is combined with error control through acknowledgements, timers and retransmission the result is an ARQ protocol. Stop-and-Wait ARQ allows a single outstanding frame, which makes it simple but leaves the channel idle for a whole round trip after every frame. Go-Back-N raises the sender window to 2^n - 1 frames while keeping the receiver window at one, so the channel stays busy, at the price of retransmitting the lost frame together with every frame that followed it. Where that penalty is too high, Selective Repeat ARQ enlarges the receiver window as well and retransmits only the damaged frame, at the cost of buffering and reordering at the receiver.

Network Topology: Definition and Ring Topology

Definition of Network Topology

Network topology refers to the physical or logical layout of a network. It defines the way different nodes are placed and interconnected with each other. Network topology also describes how the data is transferred between these nodes. Network topology is categorized into five basic models: Bus, Ring, Star, Mesh, and Tree topology.

Ring Topology

In ring topology, all network devices are connected sequentially to the same transmission line (like bus topology), except that the transmission line ends at the starting node, forming a closed loop or ring. It overcomes many of the limitations of bus topology.

Diagram

   [Node A]
   /       \
[Node D]  [Node B]
   \       /
   [Node C]

Each node is connected to exactly two other nodes, and data travels in one direction (or both directions in a dual ring) around the ring until it reaches the destination node.

Working Principle

• Data is transmitted in the form of tokens (Token Ring protocol).
• A node can only transmit data when it holds the token.
• Each node acts as a repeater, regenerating and passing the signal to the next node.
• The signal travels around the ring until it reaches the intended recipient.

Merits of Ring Topology

Demerits of Ring Topology

Summary: Ring topology provides an organized and collision-free method of data transmission using token passing, but its major weakness is that the failure of any single node or cable segment can disrupt the entire network.

Routing Table: Definition and Static vs Dynamic Routing Table

Definition of Routing Table

A routing table is a data structure stored in a router (or networked device) that lists the routes to particular network destinations. It contains information such as destination network addresses, next-hop addresses, and interface information that a router uses to determine the best path for forwarding data packets. Routers normally maintain a dynamically updating routing table based on which they make decisions on routing the data packets.

Differentiation: Static Routing Table vs Dynamic Routing Table

Summary

• Static routing is simple but inflexible; it requires manual intervention for any route change or failure.
• Dynamic routing uses protocols to discover and maintain routes automatically, making it more scalable and fault-tolerant for larger networks.

DNS System and Domain Name Space

Why Do We Need DNS When We Can Directly Use IP Addresses?

Although it is technically possible to access any resource on the internet using its IP address directly (e.g., typing 192.168.1.1 in a browser), the DNS system is needed for the following important reasons:

1. Human Readability and Memory

• IP addresses are numeric (e.g., 142.250.190.46) and are very difficult for humans to remember.
• Domain names like www.google.com are meaningful, easy to remember, and user-friendly.
• DNS acts as a translator between human-readable domain names and machine-readable IP addresses.

2. Flexibility and IP Change Management

• The IP address of a server may change over time (due to server migration, ISP changes, etc.).
• With DNS, the domain name stays the same while only the DNS record is updated in the background.
• Users never need to know or update the new IP address.

3. Load Distribution

• A single domain name can map to multiple IP addresses, allowing traffic to be distributed across several servers.
• This would not be possible if users directly used a single IP address.

4. Supporting Other Programs

• As stated in the notes, DNS is a supporting program used by other programs such as e-mail.
• Many internet services (email, FTP, web browsing) rely on DNS to function correctly.

5. Scalability

• The internet has billions of devices. Maintaining and distributing a manual list of IP-to-name mappings for every user would be impossible.
• DNS provides a distributed, hierarchical, and scalable solution to this problem.

Domain Name Space

The Domain Name Space is the hierarchical and organized structure used by DNS to manage and store all domain names on the internet.

Key Characteristics:

• The domain name space is organized as an inverted tree structure, with the root at the top.
• The tree is divided into levels, and each node in the tree has a label.
• The full domain name of any node is called its Fully Qualified Domain Name (FQDN), which is read from the node up to the root, separated by dots (.).

Structure of Domain Name Space:

            . (Root)
           / \
         com  org  net  edu  gov  (Top-Level Domains - TLD)
         /
      google
       /
     www

Example:

For the domain name www.google.com:

• com is the Top-Level Domain (TLD)
• google is the Second-Level Domain
• www is the subdomain/host

Summary:

The domain name space is a global, hierarchical, tree-structured naming system that organizes all domain names on the internet, allowing DNS to efficiently translate human-readable names into IP addresses.