Network Protocols and Packet Switching: Question 5

Worked solution

Part (a): Why a shared protocol is essential

A protocol is an agreed set of rules that every device taking part in a communication must follow. Covering things such as how the data is formatted or structured, how errors are detected and handled, and the timing of the exchange. If the sensor station and the research lab’s server did not both implement the same protocols, the server would have no reliable way to interpret the incoming data correctly: it might not recognise where one packet ends and the next begins, might not know how to check the data for errors, or might not be able to reconstruct the original file at all, so the transfer would effectively fail even if the raw bits physically reached the server.

[2 marks]: [1] for defining a protocol as a set of agreed communication rules, [1] for explaining that both devices must follow the same rules or the receiving device cannot correctly interpret/process the data.

Part (b): Packet switching via Router X and Router Y

Each of the 5 packets carries its own destination addressing and is sent independently, rather than as part of one continuous stream down a fixed path. When a packet reaches Router X, Router X examines the packet’s destination address and independently decides the most appropriate next hop for that one packet at that moment, which might be directly to Router Y, or via a different route, depending on current network conditions such as congestion or a failed link. The same independent decision is then made again at Router Y, and at any further routers, for each packet in turn.

Because this forwarding decision is made separately for every packet, and network conditions can change between one packet and the next, different packets belonging to the same file can end up travelling via different paths and taking different amounts of time to arrive. This is why the packets might not reach the research lab’s server in the order they were sent, 1, 2, 3, 4, 5, for example packet 3 might happen to take a quicker, less congested route than packet 2 and arrive first.

Once all 5 packets have arrived, the research lab’s server uses the packet numbers included with each packet to reassemble the original file in its correct order, regardless of the order the packets actually arrived in.

[4 marks]: [1] for describing packets being forwarded independently by each router based on the destination address, [1] for explaining that different packets can take different routes/times, [1] for linking this to why the arrival order can differ from the sending order, [1] for explaining that packet numbers are used to reassemble the file correctly at the destination.

Part (c): Packet switching vs circuit switching for this transfer

Benefit: packet switching lets Router X and Router Y’s links be shared between many different communications’ packets at the same time, rather than reserving the entire path exclusively for the sensor station’s transfer. A more efficient use of the network’s overall capacity, especially since a large file transfer like this does not need every part of the path constantly to itself.

Drawback: because different packets can be routed differently and can take different lengths of time, the delay between one packet and the next is not constant, and there is added processing overhead at each router (reading each packet’s header and deciding where to forward it) and at the destination (reassembling the packets into the correct order using their packet numbers).

[2 marks]: [1] for a valid benefit of packet switching over circuit switching, [1] for a valid drawback of packet switching compared with circuit switching.

Part (d): Circuit switching vs packet switching for the live video call

Circuit switching would traditionally offer the more constant and predictable data rate and delay for the whole call. It works by reserving a single, dedicated path between the two callers for the entire duration of the call, so that path is not shared with any other communication (including the sensor station’s file transfer), meaning the available bandwidth and the delay along that path stay constant throughout.

With packet switching, by contrast, the video call’s packets would share the same links as other traffic (such as the file transfer’s packets), so the delay between packets could vary and packets could even arrive with uneven gaps between them (jitter) if a link becomes busy, which is far more noticeable and disruptive in a live call than in a file transfer, where the file just needs to arrive complete eventually rather than at a constant rate.

[2 marks]: [1] for correctly identifying circuit switching, [1] for a justification based on its dedicated/reserved path giving a constant data rate and delay (or, equivalently, packet switching’s variable delay/jitter risk from sharing links with other traffic).

Final answers

• (a) Both devices must follow the same protocol (agreed rules for format, error handling and timing), or the receiving server cannot correctly interpret the data.
• (b) Router X and Router Y each independently forward every packet based on its destination address; different packets can take different routes/times and so can arrive out of order; the server uses the packet numbers to reassemble the file correctly.
• (c) Benefit: efficient shared use of the network’s links. Drawback: variable delay between packets and routing/reassembly overhead.
• (d) Circuit switching, its single dedicated, reserved path keeps the data rate and delay constant for the whole call.