Network Fundamentals →mediumMultiple Select

200-901 Network Fundamentals Practice Question

An engineer is comparing TCP and UDP for a new telemetry service that sends small, frequent datagrams from thousands of IoT sensors to a collector. The design prioritizes low overhead and does not require retransmission of lost readings. Which two characteristics apply to UDP in this scenario? (Choose two.)

⚠ Common exam trap

The trap here is assuming that a transport used for many sensors must be reliable, when UDP deliberately omits retransmission and ordering to keep overhead low for loss-tolerant data.

Answer choices

Why each option matters

Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.

Correct answer & explanation

✓

It provides no delivery guarantee, so lost datagrams are simply not retransmitted.

UDP's value in high-volume telemetry is its minimal overhead and connectionless, best-effort model. The small fixed header reduces cost per datagram, and the absence of acknowledgments and retransmissions means lost readings are simply dropped, which the design accepts. Ordering, reliability, and congestion control are all absent and would have to be added at the application layer if required.

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

✓
It provides no delivery guarantee, so lost datagrams are simply not retransmitted.
Why this is correct
UDP is a best-effort transport with no acknowledgments or retransmission timers. For thousands of sensors sending frequent small readings where the design explicitly tolerates loss, this fire-and-forget behavior eliminates the overhead and latency that TCP's reliability mechanisms would add, which matches the stated priorities.
✓
It uses a smaller header than TCP, reducing per-datagram overhead.
Why this is correct
The UDP header is a fixed eight bytes covering source port, destination port, length, and checksum, whereas TCP's header starts at twenty bytes with sequence numbers, flags, and window fields. With thousands of sensors sending many small messages, that smaller per-packet overhead meaningfully reduces bandwidth and processing cost.
✗
It guarantees in-order delivery of datagrams to the application.
Why it's wrong here
UDP makes no ordering guarantees; datagrams can arrive out of order, duplicated, or not at all. Any sequencing must be handled by the application. A collector receiving telemetry over UDP cannot assume the readings arrive in the order the sensors transmitted them, unlike a TCP stream.
✗
It performs a three-way handshake before data transmission to establish state.
Why it's wrong here
The three-way handshake is a TCP connection-establishment mechanism. UDP is connectionless and sends datagrams immediately with no prior negotiation, which is precisely why it suits a large population of sensors that transmit short, independent readings without maintaining session state with the collector.
✗
It automatically adjusts its congestion window in response to network loss.
Why it's wrong here
Congestion window management is a TCP function tied to its reliability and flow-control machinery. UDP has no congestion window, no slow start, and no backoff on loss. While this avoids TCP's throughput throttling, it also means the application must implement any pacing or congestion awareness itself.

Visual reference

Go deeper

200-901 Network Fundamentals Practice Question

Why each option matters

Option-by-option breakdown

Related to this question