M4.7 module quiz
Ten questions. 80% to pass (8 of 10). Open book, unlimited retakes.
Answers are collapsed under each question. Answer first, then check.
Q1
The four commitment instruments are:
A. RI, SP, CUD, EDP
B. Reserved, Committed, Sustained, Spot
C. Annual, triennial, monthly, hourly
D. RI, SP, CUD, Spot
Show answer
Correct: D. Spot is the asterisk: it is a pricing model rather than a commitment, and it belongs in the set because it competes for the same workloads.
Q2
effective discount = d × coverage × utilization cannot express:
A. Multi-year terms
B. A loss
C. Spot pricing
D. Multi-cloud portfolios
Show answer
Correct: B. It can never go negative, so at 40% discount and 50% utilization the truth is a 20% loss the formula cannot show. Coverage governs the benefit share; utilization against 1 − d governs win or lose.
Q3
Break-even utilization for a 40% discount is:
A. 60%
B. 40%
C. 71.4%
D. 76.9%
Show answer
Correct: A. 1 − d. Bigger discounts have looser thresholds: a 20% discount needs 80% utilization to break even.
Q4
The classic over-commitment pattern is:
A. Committing to production
B. Committing partially
C. Committing to non-prod
D. Committing for one year rather than three
Show answer
Correct: C. Non-prod is the workload most amenable to scheduling, so its utilization is the least reliable thing to commit against.
Q5
Peak-based commitment fails because:
A. Peaks are inherently hard to measure at all accurately in advance
B. Providers charge more during peaks
C. Peaks are seasonal and unpredictable
D. Utilization drops below break-even the moment the peak subsides
Show answer
Correct: D. Which is why the rule is to commit on the post-schedule floor: the floor is the only quantity that will still be running.
Q6
“Schedule first, then commit” is mathematically airtight because:
A. Schedules are simply cheaper than commitments outright, in every single case
B. Providers reward a stable scheduling history
C. Scheduling reveals the true floor, and committing above the floor strands capacity
D. Commitments cannot apply to scheduled resources
Show answer
Correct: C. The order matters because the second decision depends on the outcome of the first, and reversing it locks in a number you were about to change.
Q7
RI hoarding describes:
A. Holding the RIs centrally across all accounts
B. Refusing to sell unused RIs on
C. Buying only three-year terms
D. Buying more RIs than needed to secure capacity
Show answer
Correct: D. It converts a rate optimization into a capacity reservation nobody asked for, and the unused hours are pure loss rather than a smaller discount.
Q8
Remediating a stranded commitment usually involves:
A. Cancelling it outright
B. Exchange or sell-on, where the provider supports it
C. Reassigning it to another account entirely instead
D. Waiting for it to expire
Show answer
Correct: B. Cancellation is generally unavailable. Knowing which of exchange and marketplace resale your provider supports is what determines whether a mistake is recoverable.
Q9
A share-of-savings vendor makes sense when:
A. The org lacks the expertise and volume to manage a commitment portfolio itself
B. The organisation already has strong in-house commitment expertise itself
C. Total spend is small enough to ignore
D. The org is single-cloud already
Show answer
Correct: A. The decision math is whether their share is less than the gap between their result and what you would achieve alone, which requires an honest estimate of the second.
Q10
Auditing a share-of-savings vendor means checking:
A. Their headline total savings claim, taken entirely at face value by finance
B. The baseline they measure against and whether the savings are attributable to them
C. Their reported coverage percentage
D. Their response times to requests
Show answer
Correct: B. A vendor measuring against a baseline that ignores your own optimization claims your work as theirs, and that is the term worth negotiating before signing.
What’s next
Back to Commitments demystified.