Examkin

Question

Difficulty: MediumCost-Optimized Database Design and Capacity Planning

A university student union operates a mobile voting application on AWS for annual campus elections. The application is completely idle for 360360 days of the year, but experiences highly unpredictable, spiky traffic during a 55-day election period. The database stores simple, non-relational voting records consisting of key-value pairs. The student union operates on a very limited budget and must minimize database operational costs while ensuring the system does not throttle or fail. Which database configuration is the most cost-effective and meets these requirements?

  1. A
    An Amazon RDS for PostgreSQL Single-AZ DB instance with an active Read Replica in another Availability Zone, configured to promote the Read Replica as the primary instance to handle failover if the primary becomes overloaded.
  2. B
    An Amazon DynamoDB table configured in Provisioned capacity mode with Auto Scaling enabled, setting the minimum Write Capacity Units (WCUs) to cover baseline idle traffic and the maximum to cover the expected peak.
  3. An Amazon DynamoDB table configured in On-Demand capacity mode.Answer
  4. D
    An Amazon Aurora PostgreSQL DB cluster with a provisioned writer instance and two reader instances, utilizing an Aurora Auto Scaling policy to add reader instances during peak hours.

Answer

An Amazon DynamoDB table configured in On-Demand capacity mode is the most cost-effective solution.
The correct answer is the option specifying Amazon DynamoDB in On-Demand capacity mode. Because the application is idle for 360360 days a year, On-Demand capacity mode ensures there are no hourly write or read capacity charges during the idle period. During the 55-day election period, it instantly scales to accommodate any sudden write spikes, providing zero-throttling reliability at the lowest possible cost.

Step-by-Step Solution

1
Analyze the application workload characteristics, traffic patterns, and data model.
The database workload is non-relational key-value data, completely idle for 360360 days of the year, and subject to sudden, unpredictable traffic spikes during a 55-day window.
Identifying the workload and data structure helps determine whether a relational or non-relational database is appropriate and informs the scaling strategy.
2
Select the correct database engine based on data model and baseline costs.
Amazon DynamoDB is selected over relational engines (RDS and Aurora) because it natively supports key-value storage and avoids the continuous hourly compute costs of idle virtual instances.
Eliminating relational database options avoids unnecessary baseline compute costs during the long idle periods.
3
Evaluate the capacity planning options for the chosen database engine to optimize cost and performance.
DynamoDB On-Demand capacity mode is chosen instead of Provisioned capacity mode because On-Demand scales instantly to handle unpredictable peaks without throttling, and charges only for requests consumed, resulting in near-zero cost during idle periods.
Choosing On-Demand capacity mode aligns database costs directly with actual usage, which is ideal for spiky, unpredictable, and primarily idle workloads.

Key Concept

Selecting the cost-optimal database engine and capacity scaling mode for spiky, highly irregular workloads.
Estimated Time:1m 30s
Rate this question

Topics

Cost-Optimized Database Design and Capacity Planning