KCNA Cloud Native Observability Practice Question

Answer analysis

Option-by-option breakdown

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

• ✓

  rate()

  Why this is correct

  rate() computes the per-second average rate of increase of a counter over the given range window, automatically handling counter resets. It directly satisfies the stem's requirement to derive per-second increase from a monotonically increasing counter, unlike irate() or increase().

• ✗

  delta()

  Why it's wrong here

  delta() computes the difference between the first and last sample values in the range, intended for gauges. Applied to a counter it ignores resets and yields no per-second normalisation. delta() is correct for tracking how much a gauge, such as temperature or queue depth, changed.

• ✗

  avg_over_time()

  Why it's wrong here

  avg_over_time() averages the sample values within the range, producing a smoothed gauge value rather than a rate. It performs no counter-reset handling or division by window seconds. avg_over_time() is correct for smoothing noisy gauges such as CPU utilisation or memory usage.

• ✗

  increase()

  Why it's wrong here

  increase() returns the total counter growth across the window as an absolute value, not divided by the window duration. Per-second rate requires rate(), which divides the increase by the range in seconds. increase() suits dashboards showing how many events occurred over a period.