Q.1
14 MarksA Computer Network is an interconnection of two or more autonomous computing devices linked by communication media (cables, fiber, wireless) and protocols to exchange data and share hardware/software resources.
Primary Types of Computer Networks:
- PAN (Personal Area Network): Covers personal workspace within (e.g., Bluetooth connection between smartphone and smartwatch).
- LAN (Local Area Network): Spans a single room, office, or university campus () with very high data rates of (e.g., College lab Ethernet/Wi-Fi).
- MAN (Metropolitan Area Network): Covers an entire town or city (), typically operated by cable TV providers or municipal utilities.
- WAN (Wide Area Network): Spans countries and continents using telecom fiber backbones and satellite links (e.g., The global Internet).
Also asked in Summer 2026 Q3b OR, Winter 2025 Q1b, Winter 2024 Q1b
When a data packet travels across a packet-switched network from source to destination, it encounters four distinct delay components at each intermediate router hop:
- Processing Delay ():
- The time taken by the router CPU to inspect the packet header, verify the bit-level checksum for transmission errors, and determine the output link interface via routing table lookup.
- Typical Value: Microseconds ().
- Queuing Delay ():
- The duration a packet waits in the router output buffer queue before being scheduled for transmission onto the physical link.
- Highly dependent on traffic intensity (). If buffer is full, packet drop occurs.
- Typical Value: Microseconds to milliseconds.
- Transmission Delay ():
- The time required to push (pump) all packet bits onto the communication medium.
- Formula:
where = packet length in bits, and = transmission link rate (bandwidth) in bits per second (bps).
- Propagation Delay ():
- The time taken for a physical bit to travel from the start of the physical wire to the destination router interface.
- Formula:
where = physical distance between routers, and = propagation speed of light in the medium (approx. in copper/fiber).
Also asked in Winter 2024 Q1c
A Network Topology defines the geometric and structural layout of communication links and nodes in a computer network. Topologies are categorized as Physical (actual cable routing) or Logical (data transmission path).
1. Star Topology
- Design: All devices connect directly to a central networking hub, switch, or router via dedicated point-to-point links.
- Advantages: Easy installation and maintenance; single device failure does not affect the rest of the network; easy fault isolation.
- Disadvantages: Single point of failure (if central switch fails, entire network crashes); higher cabling requirements than bus.
2. Bus Topology
- Design: All devices share a single linear backbone communication cable (the "bus"). Terminators are attached at both ends to absorb signal reflection.
- Advantages: Minimal cabling required; cost-effective for small networks.
- Disadvantages: Backbone cable break halts the whole network; heavy packet collisions under high traffic; difficult to isolate faults.
3. Ring Topology
- Design: Devices form a continuous circular loop. Each node has exactly two neighbors. Data circulates sequentially using Token Passing.
- Advantages: Orderly network access with zero packet collisions due to token passing; performs well under heavy load.
- Disadvantages: Unidirectional ring fails completely if any single node or cable breaks; high transmission latency for large loops.
4. Mesh Topology
- Design: Every device is connected to every other device via dedicated point-to-point channels.
- Advantages: Maximum fault tolerance and redundancy; no traffic congestion; excellent data privacy and security.
- Disadvantages: Extremely expensive cabling and complex I/O hardware port requirements.
5. Tree (Hierarchical) Topology
- Design: Hierarchical star networks branched off a central root bus or master switch.
- Advantages: High scalability; easy management and departmental isolation.
- Disadvantages: Root switch dependency; heavy cabling.
6. Hybrid Topology
- Combines two or more basic topologies (e.g., Star-Ring or Star-Bus) to balance cost, reliability, and coverage in large enterprise networks.