What is 'Store-and-Forward' transmission?

It is a networking technique where a switch must receive the entire packet before it can begin transmitting the first bit of that packet onto the next outgoing link.

What is the primary difference between how resources are allocated in circuit switching versus packet switching?

Circuit switching uses dedicated resource reservation where a specific path is locked for one user. Packet switching uses on-demand allocation where resources are shared among all users via statistical multiplexing.

When is circuit switching more efficient than packet switching?

It is more efficient for long-duration, constant-bit-rate applications like traditional voice calls. In these cases, the overhead of the initial setup is offset by the lack of per-packet header overhead and zero queuing delay.

How does the handling of network congestion differ between the two switching methods?

In circuit switching, congestion happens at the start; if no path is available, the connection is blocked (busy signal). In packet switching, congestion happens during transit, leading to packets being stored in buffers (queuing delay) or dropped if buffers are full.

What is a common misconception regarding the 'speed' of packet switching?

Students often think packet switching is 'faster' because it's modern. In reality, a single packet-switched link may have higher latency due to store-and-forward delays; its advantage is efficiency and capacity, not necessarily raw per-packet speed.

Why is it incorrect to say packet switching has no delay?

Packet switching introduces several delays, most notably transmission delay ($L/R$) and queuing delay. While it avoids the setup delay of circuit switching, the cumulative nodal delays can be significant under high load.

What error is made when calculating transmission delay using the distance of the cable?

Transmission delay is calculated as $L/R$ (Packet length / Link rate) and is independent of distance. Distance is only used to calculate propagation delay ($d/s$), which is the time for a bit to travel the physical length of the medium.

Define 'Statistical Multiplexing' in the context of packet switching.

It is a method of sharing a communication channel where the bandwidth is divided into a variable number of channels based on the actual demand of the users, rather than fixed pre-allocated slots.

What is the 'Header' of a packet and why is it necessary?

The header is a block of control information at the start of a packet. It is necessary because it contains the destination address, allowing switches to route the packet independently without a pre-established path.

Why does packet switching allow more users to share a network than circuit switching?

Because users rarely transmit at their peak rate constantly. Packet switching exploits these idle periods to send data for other users, whereas circuit switching leaves the capacity unused if the dedicated user is silent.

Packet & Circuit Switching | OCR AS-Level Computer Science

Packet & Circuit Switching

Summary

Network switching defines how data is moved across a network of interconnected nodes. The two primary paradigms are Circuit Switching, which relies on dedicated end-to-end resource reservation, and Packet Switching, which utilizes shared resources and discrete data units to maximize efficiency and resilience.

1. Definition & Core Concepts

Circuit Switching is a methodology where a dedicated physical or logical communication path is established between two nodes before data transfer begins. This path remains reserved for the exclusive use of the connection until it is explicitly terminated.
Packet Switching involves breaking down a message into smaller, independent units called packets. Each packet contains a header with routing information and is transmitted across the network using shared resources, potentially taking different paths to the destination.
The primary goal of switching is to manage how multiple users share a limited set of network links and switches without requiring a direct physical wire between every possible pair of users.

Comparison of Circuit Switching showing a solid dedicated path versus Packet Switching showing discrete packets on a shared path.

2. Underlying Principles

Resource Reservation: In circuit switching, bandwidth is reserved at each link along the path. This guarantees a constant bit rate and zero queuing delay once the connection is established, but leads to waste if the line is idle.
Statistical Multiplexing: Packet switching uses this principle to share link capacity. Resources are allocated on-demand, allowing multiple users to utilize the same link simultaneously based on the statistical likelihood that not everyone is sending data at the exact same moment.
Store-and-Forward: Packet switches must receive the entire packet before they can begin transmitting the first bit onto the next link. This introduces a transmission delay proportional to the packet size and link speed.

3. Methods & Techniques

Connection Setup: Circuit switching requires a three-phase process: setup (signaling), data transfer, and teardown. This overhead makes it less efficient for short, bursty data transfers.
Routing and Forwarding: In packet switching, each switch uses a forwarding table to determine the next hop for a packet based on its destination address. This allows the network to dynamically route around failures.
Packet Encapsulation: Data is wrapped in headers containing source/destination addresses, sequence numbers for reassembly, and checksums for error detection. This overhead is necessary for the independent delivery of data units.

4. Key Distinctions

5. Exam Strategy & Tips