This set builds vocabulary for simulating traffic and measuring system performance under load.
0 / 5 completed
1 / 5
At standup, a dev proposes simulating many concurrent virtual users hitting an API to measure how it behaves under heavy traffic before a launch. What is this practice called?
Load testing simulates realistic or heavy traffic against a system to observe how latency, error rates, and resource usage behave under that load, revealing issues that wouldn't surface in low-traffic manual testing. This is especially important before high-stakes events like a product launch or marketing campaign. It differs from unit testing, which verifies logic correctness rather than performance under load.
2 / 5
During a design review, the team wants to gradually increase simulated traffic to find the exact point where response times start degrading. What is this testing pattern called?
A ramping or stress test gradually increases simulated load over time, letting the team identify the specific traffic level at which performance starts to degrade or the system fails outright. This pinpoints capacity limits more precisely than a single fixed-load test. Knowing this breaking point informs both capacity planning and scaling configuration.
3 / 5
In a code review, a dev defines a threshold that automatically fails the test run if the 95th percentile response time exceeds a target value. What is this called?
A performance threshold defines an automated pass/fail condition, such as p95 latency staying under a target value, turning a load test into an objective gate that can be integrated into CI rather than requiring manual result interpretation each time. This automation makes performance regressions catchable before they reach production. Defining meaningful thresholds requires understanding the system's actual performance requirements.
4 / 5
An incident report shows a service that passed all load tests still failed under real production traffic due to a different, more bursty traffic pattern. What testing gap does this reveal?
If a load test's simulated traffic pattern, such as steady ramp-up, doesn't match the real-world pattern, like sudden bursts, the test can pass while still missing a failure mode that only appears under actual production conditions. Designing realistic traffic scenarios, not just steady load, is important for meaningful load testing. This gap is a common lesson learned when a service fails despite passing prior load tests.
5 / 5
During a PR review, a teammate asks how a scripted load test using virtual users differs from simply monitoring real production traffic after launch. What is the key advantage?
Load testing with virtual users lets a team discover performance problems in a controlled, repeatable setting before real users are ever exposed to them, whereas monitoring production traffic only reveals issues after they've already affected live users. Both practices are valuable and complementary, but load testing is specifically about proactive discovery ahead of launch. This proactive advantage is why load testing is a standard pre-launch practice.