AWS Certified Solutions Architect – Associate (SAA-C03)Design High-Performing ArchitecturesEasy
A research institution needs to perform complex scientific simulations that require massive parallel processing capabilities. These simulations can run for hours or days, are computationally intensive, and can tolerate interruptions. The institution wants to minimize costs while ensuring access to powerful compute resources.
- AUse On-Demand EC2 instances with Auto Scaling.
- BRun simulations on AWS Lambda functions.
- CProvision EC2 Dedicated Hosts.
- DUtilize EC2 Spot Instances within an Auto Scaling group.
Show answer & explanationAnswer & explanation
Correct answer: D. Utilize EC2 Spot Instances within an Auto Scaling group.
EC2 Spot Instances allow you to bid on unused EC2 capacity, offering significant cost savings (up to 90% off On-Demand prices). Since the scientific simulations can tolerate interruptions and are computationally intensive, Spot Instances are an ideal, cost-effective solution for massive parallel processing, especially when managed by an Auto Scaling group for resilience.
Why the other options are wrong
- A. On-Demand instances are reliable but significantly more expensive than Spot Instances, which is not ideal for cost minimization for interruptible workloads.
- B. Lambda functions are for short-duration, event-driven tasks (max 15 minutes) and are not suitable for 'simulations that can run for hours or days' or complex scientific computing requiring specific instance types.
- C. Dedicated Hosts provide physical EC2 servers for specific licensing needs and are very expensive; they do not align with cost minimization for interruptible workloads.
EC2 Spot Instances
Amazon EC2 Spot Instances let you take advantage of unused EC2 capacity in the AWS cloud, available at a discount of up to 90% compared to On-Demand prices.
- Ideal for fault-tolerant, flexible, and stateless workloads.
- Can be interrupted by AWS with a two-minute warning.
- Significant cost savings compared to On-Demand instances.
- Often used for big data, containerized workloads, CI/CD, and HPC.
Memory trick: Spot Instances: Savings for Spiky, Stop-able tasks.