Difficulty
Calculator
Interactive practice · 25 questions on bandwidth vs throughput, latency, how to calculate and optimize each, common bottlenecks, key performance metrics, and the tools that measure them — with step-by-step explanations.
Press Check answer on each question to see whether you got it right — every check is timestamped and recorded in your Activity log below.
Each attempt is saved under its own name — load, rename, or delete them below. Difficulty / calculator tags are shared across all attempts. Numeric answers accept equivalent forms (e.g. 4/3 or 1.333).
The timer is stopped. Questions are hidden until you resume.
Easy Which pair of definitions is correct?
Answer: D — Bandwidth = the maximum capacity of a link (e.g. 1 Gbps); throughput = the actual data rate achieved over it
Easy For a given link, how do throughput and bandwidth relate?
Answer: B — Throughput ≤ bandwidth (throughput can never exceed the link's capacity)
Medium Your ISP sells a 100 Mbps plan, but your download manager shows about 12.5 MB/s. Why?
Answer: A — Mbps is megabits/sec and MB/s is megabytes/sec — divide by 8: 100 ÷ 8 = 12.5
100 Mbps ÷ 8 = 12.5 MB/s — so 12.5 MB/s is exactly what a 100 Mbps link delivers.Medium At its theoretical maximum, how many seconds does it take to download a 100 MB file over a 100 Mbps link?
seconds
Answer: 8
100 MB × 8 = 800 megabits.800 Mb ÷ 100 Mbps = 8 seconds (real transfers are a bit longer due to overhead).Medium Why is real throughput usually lower than the link's bandwidth? Select all that apply.
Answer: A, C, D
Easy What is the difference between latency and throughput?
Answer: C — Latency = time for one unit of work to complete; throughput = number of units completed per unit time
Medium You double a link's bandwidth. What happens to a single request's latency?
Answer: B — Latency is largely unchanged — a wider pipe raises throughput, but distance/round-trips still dominate one request's time
Hard By Little's Law, throughput = concurrency ÷ latency. If a service holds 50 requests in flight and average latency is 200 ms, what is its throughput in requests/second?
req/s
Answer: 250
50 / 0.200 s = 250 requests/second. (Raising concurrency or cutting latency raises throughput.)Hard The bandwidth-delay product for a 100 Mbps link with an 80 ms round-trip time is:
Answer: B — 1 MB — the amount of data 'in flight' the link can hold (100 Mbps × 0.08 s = 8 Mb = 1 MB)
Medium Total network latency for a packet is the sum of which delays? Select all that apply.
Answer: A, B, C, D
Medium Transmission delay = packet size ÷ bandwidth. For a 1500-byte packet on a 1 Gbps link, it is about:
Answer: D — 12 microseconds (1500 × 8 bits ÷ 1e9 bps = 12 µs)
1500 bytes × 8 = 12,000 bits.12,000 ÷ 1,000,000,000 = 1.2e-5 s = 12 µs.Medium Which techniques primarily reduce latency? Select all that apply.
Answer: A, B, D
Medium Which techniques primarily increase throughput? Select all that apply.
Answer: A, B, C, D
Hard An API is slow because the client makes 30 sequential calls, each a full round trip. The best fix is:
Answer: D — Reduce round trips — batch the calls into one request (or fetch in parallel)
Medium Bufferbloat (oversized network buffers) causes which problem?
Answer: D — High latency under load — packets sit in long queues instead of being dropped early
Medium Head-of-line blocking is when:
Answer: A — One slow/stuck item at the front of a queue holds up all the items behind it
Hard As you increase load, throughput plateaus while latency climbs sharply. This means:
Answer: D — A resource is saturated (the bottleneck) — requests queue instead of being served faster
Medium Why do teams report p95 / p99 latency instead of the average?
Answer: D — Averages hide the slow tail — p99 shows the experience of your worst-affected 1% of requests
Medium Which are core system performance metrics worth tracking? Select all that apply.
Answer: A, B, C, D
Medium Which tool measures raw bandwidth/throughput between two hosts?
Answer: C — iperf (or iperf3)
iperf pushes traffic end-to-end and reports achievable throughput. ping/traceroute measure latency and path, not bandwidth.Easy Which tool measures round-trip latency to a host?
Answer: A — ping
ping sends ICMP echoes and reports round-trip time (and jitter/loss). traceroute/mtr add per-hop latency.Medium Which of these are load-testing tools for measuring throughput/latency under traffic? Select all that apply.
Answer: A, B, C, D
Medium Which stack is commonly used to collect metrics and build dashboards?
Answer: D — Prometheus (metrics) + Grafana (dashboards)
Medium Jitter refers to:
Answer: C — Variation (inconsistency) in latency between packets — especially harmful for VoIP/video
Hard Goodput is best defined as:
Answer: C — The rate of useful application data delivered — throughput minus protocol overhead and retransmissions
goodput ≤ throughput ≤ bandwidth. It's what the user actually experiences.