A company is migrating a database workload to AWS. The database experiences sudden, unpredictable spikes in traffic that last for only a few minutes, followed by long periods of idle time. The solutions architect needs to select a database capacity strategy that prevents application throttling while minimizing costs for idle resources. Which configuration meets these requirements most cost-effectively?
- Amazon DynamoDB configured with On-Demand capacity modeAnswer
- BAmazon DynamoDB configured with Provisioned capacity mode and a high constant throughput limit
- CAmazon DynamoDB configured with Provisioned capacity mode and Application Auto Scaling
- DAmazon RDS Multi-AZ deployment utilizing Read Replicas as the primary disaster recovery failover target to handle the write spikes
Answer
Amazon DynamoDB configured with On-Demand capacity mode
The configuration utilizing Amazon DynamoDB configured with On-Demand capacity mode is correct. On-demand capacity mode is ideal for highly unpredictable or spiky workloads because it instantly accommodates the traffic peaks and charges only for the reads and writes performed, avoiding costs during idle hours.
Step-by-Step Solution
Key Concept
Selecting between DynamoDB On-Demand and Provisioned capacity modes based on workload predictability and spikiness to optimize costs.