INFRASTRUCTURE LEARNINGINFRASTRUCTURE LEARNING
インフラ大全インフラ大全IT INFRASTRUCTURE / LEARNING
Infrastructure learning guide
Find a lesson or lab/

Esc to close · Search terms stay on your device

LEARNING日本語

Topic lessonTopic: Applied GitHub Actions and SQL Server recovery labs

CloudWatch alarm exercise|Threshold and missing data

Static thresholds use evaluation periods, required breaching points and a missing-data policy. A missing value differs from an observed zero.

What you will learn

Compare missing-data policies in a two-of-three teaching model.

Prerequisite

The cases use fictional inputs.

Start the exercise →

What to predict

Static thresholds use evaluation periods, required breaching points and a missing-data policy. A missing value differs from an observed zero.

From conditions to the decision
From conditions to the decision1: Point 1 — Align conditions — Point 2. 2: Point 2 — Compare outcomes — Point 3. 3: Point 3 — Verify the decision — Missing-data policy123Point 1Point 2Point 3Missing-data policyFrom conditions to the decision1: Point 1 — Align conditions — Point 2. 2: Point 2 — Compare outcomes — Point 3. 3: Point 3 — Verify the decision — Missing-data policy123Point 1Point 2Point 3Missing-datapolicy
  1. Point 1 → Point 2Align conditions
  2. Point 2 → Point 3Compare outcomes
  3. Point 3 → Missing-data policyVerify the decision
Point 1
90 / 10
Point 2
20 / 95
Point 3
Missing / 15
Missing-data policy
Keep unknown / Treat as breaching

Compare the cases

Synthetic observations
CaseConditionsDecision
Trial 1Point 1 90, Point 2 20, Point 3 Missing, Missing-data policy Keep unknownInsufficient evidence: 1 known breaches, 1 missing points.
Trial 2Point 1 90, Point 2 95, Point 3 Missing, Missing-data policy Keep unknownALARM: 2 points count as breaches, with 1 missing.
Trial 3Point 1 90, Point 2 20, Point 3 15, Missing-data policy Keep unknownOK: 1 points count as breaches, with 0 missing.
Trial 4Point 1 90, Point 2 20, Point 3 Missing, Missing-data policy Treat as breachingALARM: 2 points count as breaches, with 1 missing.

Acceptance and limits

A fixed model alarms when two of three values meet 80. It omits extra evaluation ranges, retained state and anomaly detection.

Read the initial conditions and operational goal.

Change one condition and compare the decision.

Record the model boundary and outstanding checks.

Change a condition

A fixed model alarms when two of three values meet 80. It omits extra evaluation ranges, retained state and anomaly detection.

Insufficient evidence: 1 known breaches, 1 missing points.

Check your understanding

Choose an option to see the result and explanation.

What happens in trial 1?
Read the answer and reason

Answer: Insufficient evidence: 1 known breaches, 1 missing points. · Point 1 90, Point 2 20, Point 3 Missing, Missing-data policy Keep unknown → Insufficient evidence: 1 known breaches, 1 missing points.

What happens in trial 2?
Read the answer and reason

Answer: ALARM: 2 points count as breaches, with 1 missing. · Point 1 90, Point 2 95, Point 3 Missing, Missing-data policy Keep unknown → ALARM: 2 points count as breaches, with 1 missing.

What happens in trial 3?
Read the answer and reason

Answer: OK: 1 points count as breaches, with 0 missing. · Point 1 90, Point 2 20, Point 3 15, Missing-data policy Keep unknown → OK: 1 points count as breaches, with 0 missing.

Check, then explain in your own words

  1. Read the initial conditions and operational goal.
  2. Change one condition and compare the decision.
  3. Record the model boundary and outstanding checks.
Check your understanding: Compare missing-data policies in a two-of-three teaching model.

A fixed model alarms when two of three values meet 80. It omits extra evaluation ranges, retained state and anomaly detection.

Record the model boundary and outstanding checks.

Sources and conditions checked

The official references used in this lesson are collected here with their scope, conditions and corresponding sections. Independent expert review and live AWS comparison have not been performed.