Question #1319
A manufacturing firm must migrate a legacy monitoring system from an on-premises environment to AWS due to increasing operational costs. The system operates continuously and generates a steadily expanding dataset from real-time sensor inputs. What should a solutions architect recommend to address these requirements MOST cost-effectively?
Migrate the application layer to Amazon EC2 Spot Instances. Migrate the dataset to Amazon DynamoDB with on-demand capacity.
Migrate the application layer to Amazon EC2 On-Demand Instances. Migrate the dataset to Amazon RDS with on-demand instances.
Migrate the application layer to Amazon EC2 Reserved Instances. Migrate the dataset to Amazon Aurora Reserved Instances.
Migrate the application layer to Amazon EC2 Reserved Instances. Migrate the dataset to Amazon S3 Intelligent-Tiering.
Explanation
Option C is correct because:
- EC2 Reserved Instances: The system operates continuously, making RIs ideal due to their lower cost compared to On-Demand/Spot Instances. RIs suit predictable, long-term workloads.
- Aurora Reserved Instances: Aurora offers MySQL/PostgreSQL compatibility and scalability. Using RIs for Aurora reduces costs for steady database workloads, which aligns with the expanding dataset from real-time sensors.
Other options are incorrect because:
- A: Spot Instances are unreliable for continuous systems, and DynamoDB on-demand may cost more than RIs for steady data growth.
- B: On-Demand pricing for EC2/RDS is more expensive than RIs for long-term use.
- D: While S3 Intelligent-Tiering is cost-effective for storage, it’s unsuitable for real-time database queries required by monitoring systems. Aurora RIs better support real-time data processing.
Key Points: Use RIs for predictable, continuous workloads; Aurora suits real-time database needs; avoid Spot/On-Demand for steady-state systems.
Answer
The correct answer is: C