Objective: distinguish fit from a price claim
After lessons 01 and 07, use interruption tolerance and commitment conditions to explain EC2 purchasing choices. On-Demand is usage-based purchasing. Spot suits work that is flexible about when it runs and can tolerate interruption. Savings Plans involve a usage-amount commitment, while Reserved Instances involve instance-configuration commitments. These are different commitment dimensions, not interchangeable labels. A conceptual fit is not a current rate quote, a guaranteed capacity allocation, or proof that an option is cheapest for every workload.
Worked comparison: two simulation tasks
Fictional teaching case: a simulation project has a resumable overnight batch with flexible completion timing and a separate interactive demonstration whose use is unpredictable. The batch can save progress, so interruption is a stated tolerated condition; Spot is an option to evaluate for that component. The demonstration has no established sustained commitment and is not described as interruption-tolerant; On-Demand is an appropriate category to evaluate. First separate the tasks, then preserve their different conditions. Check the reasoning by removing the batch’s ability to resume: the original Spot rationale would need reconsideration. Neither selection supplies actual prices, savings percentages, or a completed estimate.
Worked subtotal with wholly hypothetical rates
For arithmetic only, suppose three imaginary instances run for 12 hours at a fictional $0.10 per instance-hour. Suppose 50 GB of storage costs a fictional $0.02 per GB for that same stated period. These are invented worksheet rates, not AWS prices. Compute usage is 3 × 12 = 36 instance-hours, so its component is $3.60. Storage is 50 × $0.02 = $1.00 because its rate already covers the given period. Add like monetary components to obtain a $4.60 supplied-data subtotal. Check units before adding: instance-hours cancel the compute rate, while the storage rate cancels GB for the specified period. Do not multiply a period rate by an extra count of days.
Why a correct subtotal is still partial
A real environment may include components beyond compute, such as EBS snapshot storage. Estimates need relevant configuration quantities and applicable pricing conditions; variable demand can change the inputs. Pricing Calculator models the information entered and excludes applicable taxes. A correct arithmetic subtotal therefore can coexist with an incomplete cost model. The honest response is to label included components and identify missing applicable inputs, not to fill gaps with remembered rates. Also separate billing analysis from choosing a purchasing category: neither one independently establishes the full cost of the other.
Transfer check and check basis
Try this check: a fictional worksheet supplies storage at $0.04 per GB for the entire stated week and 20 GB is stored. A learner multiplies by seven again. Is that justified? Check basis: not under the supplied units; 20 × $0.04 = $0.80 already covers that week. Now explain why adding this correct storage amount to compute does not prove a final AWS invoice. The missing model boundary still matters. Use aws-clf-q-041–044 and cards aws-clf-card-31–33, stating the unit calculation and every omitted component rather than reporting an unsupported universal saving.
Try it, then explain it.
Complete a short set without looking at the example. After each answer, explain which evidence or rule supports your decision and why the alternatives do not fit.
Sources & authorship
Original OpenExam explanations and examples. References support the concepts; they are not the source of copied questions. Independent educator review is pending.
What is Amazon EC2? - Amazon Elastic Compute CloudAmazon EC2 billing and purchasing options - Amazon Elastic Compute CloudWhat is AWS Pricing Calculator? - AWS Pricing Calculator