Archera defines commitment coverage a bit differently from AWS to make the evaluation process for Commitment Strategies to be as intuitive as possible.

While AWS defines Commitment coverage as the total number of On-demand Running Hours being covered by Commitments, Archera defines Commitment coverage as the total On-demand Spend in $'s being covered by Commitments.

This makes it more clear to customers what costs are being committed to instead of resource hours that can be billed at wildly different rates.

For example if you are running both a t3, cheap compute VM, and a p4, expensive GPU VM, for 24 hours a day on-demand you would pay $5 a day for the t3 & $770 a day for the P4 for a total of $775 a day with 0% coverage.

According to AWS if you buy a commitment for the t2 you increase coverage to 50% but are still paying $772 a day.

According to Archera if you buy a commitment for the t2 you increase coverage to only 1% since the majority of your $772 a day spend is On-Demand.

This gives a consistent metric to compare between Commitment strategies and also helps avoid strange cases of plans with much lower coverage yielding far higher saving compare to plans covering many more cheaper machines.

Did this answer your question?