Question on regarding Spot instance…

[lijiann]

A company is modernizing its on-premises system by migrating it to AWS. The system will be hosted on EC2 instances managed by Amazon Elastic Kubernetes Service (EKS) and use Amazon RDS for MySQL as the database. The system has predictable schedules of high usage, especially during sales events and holiday seasons.

What pricing options should the company consider when selecting the MOST cost-optimized solution?

Purchase Compute Savings Plans for the EC2 nodes of the EKS cluster to be used for regular traffic. Scale the node cluster with Spot instances during peak demands. Handle the predicted database load with a 1-year All Upfront Reserved Instance and vertically scale up the DB instance on scheduled high usage.

Exam is saying this is the correct answer with the Spot instance. I get the savings part, but where in the question state the he workload is flexible and can handle interruptions, to make this a suitable candidate for Spot instance?

• This discussion was modified 9 months, 1 week ago by  lijiann.

[lijiann]

i guess the later questions answers this… diversify spot instances…

[segan]

This STILL does not make sense since the answer choice never mentions anything about using a spot fleet where scheduled scaling could be setup. It only says using “spot instances”.

[JR-TutorialsDojo]

Hello lijiann,

Thank you for the feedback.

The scenario describes a system with predictable periods of high usage, particularly during sales events and holiday seasons. In such cases, Spot Instances are a suitable solution for handling temporary and high-capacity compute demands in a cost-effective manner. They are ideal when the extra capacity required is not constant but occurs in scheduled bursts, allowing the company to take advantage of lower-cost, short-term compute resources. This is especially effective when the base workload is already handled by a more stable and committed pricing option, such as Compute Savings Plans or Reserved Instances.

For more information, please refer to these:
https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/spot-fleet-scheduled-scaling.html
https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/using-spot-instances.html

I hope this helps! Let us know if you need further assistance.

Regards,
JR @ Tutorials Dojo

[segan]

This STILL does not make sense since the answer choice never mentions anything about using a spot fleet where scheduled scaling could be setup. It only says using “spot instances”.

• This reply was modified 2 months, 1 week ago by  segan.

[JR-TutorialsDojo]

Hello segan,

Thank you for sharing your thoughts on this item.

The question specifically asks about the pricing options the company should consider, not the exact implementation details of scaling. That’s why the answer choice simply mentions “Spot instances.” It doesn’t need to explicitly say “Spot Fleet,” because in practice, Spot scaling in Amazon EKS is typically implemented through Auto Scaling groups with Spot capacity pools, not necessarily Spot Fleet. The key point is that Spot capacity provides the most cost‑optimized way to handle predictable bursts in demand compared to On‑Demand or Dedicated Instances.

For more information, you may refer to this: https://docs.aws.amazon.com/eks/latest/best-practices/cas.html

I hope this helps! Let us know if you need further assistance.

Regards,
JR @ Tutorials Dojo