Examkin

Question

Difficulty: HardAWS Pricing Models

An online education platform is migrating its core infrastructure to Amazon EC2. The platform's workload profile includes:

* An administrative student portal database that must run continuously 24/7 with a highly predictable baseline compute load.
* A video transcoding service that processes user-uploaded lecture videos; this process is stateless, fault-tolerant, can be paused or resumed, and has no strict completion deadline.
* A virtual classroom application that experiences highly unpredictable, short-duration spikes in user traffic whenever a popular live session is announced.

To optimize costs while meeting application requirements, which of the following pricing model strategies should the company implement? (Select TWO).

  1. Deploy the video transcoding service using Spot Instances to utilize spare capacity at the lowest price point.Answer
  2. Deploy the virtual classroom application's dynamic scaling capacity using On-Demand Instances to accommodate unpredictable traffic spikes.Answer
  3. C
    Deploy the administrative student portal database using Spot Instances to run the persistent database at a discount.
  4. D
    Deploy the virtual classroom application using Dedicated Hosts to automatically scale down to zero when there is no user demand.
  5. E
    Deploy the administrative student portal database using On-Demand Instances to cover the long-term, predictable 24/7 baseline load.

Answer

Deploy the video transcoding service using Spot Instances and deploy the virtual classroom application's dynamic scaling capacity using On-Demand Instances.
The correct strategy uses Spot Instances for the video transcoding service because it is a stateless, fault-tolerant workload that can handle interruptions, allowing the company to run it at a significant discount. It also uses On-Demand Instances for the virtual classroom application's dynamic scaling capacity because the traffic spikes are unpredictable, meaning the company needs immediate, reliable compute capacity without committing to long-term costs.

Step-by-Step Solution

1
Analyze the video transcoding workload characteristics.
The video transcoding job is identified as stateless, fault-tolerant, and flexible with no strict completion deadline.
Workloads with these traits are optimal candidates for Spot Instances, which provide spare AWS capacity at significant discounts in exchange for potential interruption.
2
Analyze the virtual classroom workload characteristics.
The virtual classroom application experiences highly unpredictable, short-duration spikes.
Unpredictable spikes require instant scale-up capabilities without any long-term capacity commitment, making On-Demand Instances the most appropriate choice.
3
Analyze the student portal database workload characteristics.
The database requires continuous 24/7 operation with a highly predictable baseline compute load.
Predictable, steady-state workloads should be covered by Reserved Instances or Savings Plans to secure discounts, rather than expensive On-Demand pricing or interruptible Spot Instances.

Key Concept

Selecting the optimal Amazon EC2 pricing model based on workload characteristics (predictability, fault tolerance, and duration) to maximize cost efficiency.
Rate this question

Topics

AWS Pricing Models

More questions from this topic