A company is optimizing the costs of its database tier on AWS. The current architecture co… — AWS Certified Solutions Architect - Associate

A company is optimizing the costs of its database tier on AWS. The current architecture consists of:

* An Amazon RDS for PostgreSQL database that requires $200\text{ GB}$ of storage. The database handles OLTP workloads with a constant baseline of $3,000\text{ IOPS}$ , but experiences daily performance spikes up to $9,000\text{ IOPS}$ .
* An Amazon DynamoDB table used to store historical user activity logs. The log generation is highly unpredictable, experiencing sudden write bursts of up to $2,000\text{ writes/sec}$ followed by hours of complete inactivity.

Which combination of database configurations will meet these requirements most cost-effectively? (Select TWO.)

Configure the RDS for PostgreSQL DB instance to use gp3 storage with $9,000$ provisioned IOPS.Answer
Configure the DynamoDB table to use On-Demand capacity mode.Answer
C
Configure the DynamoDB table to use Provisioned capacity mode with Auto Scaling enabled and a target utilization of $70\%$ .
D
Configure the DynamoDB table to use Provisioned capacity mode with a fixed baseline of $2,000$ Write Capacity Units (WCUs) to handle peak write bursts.
E
Configure the RDS for PostgreSQL DB instance as a Single-AZ instance and deploy a Read Replica in another Availability Zone to serve as the automatic failover target.

Answer

The correct database configurations are to configure the RDS for PostgreSQL DB instance to use gp3 storage with 9,000 provisioned IOPS, and to configure the DynamoDB table to use On-Demand capacity mode.

Configuring the RDS for PostgreSQL DB instance to use gp3 storage with

9,000

provisioned IOPS is correct because gp3 allows independent provisioning of IOPS and storage volume size. This avoids the cost of scaling a gp2 volume to

3,000\text{ GB}

just to get

9,000\text{ IOPS}

, and is more cost-effective than io2 storage. Configuring the DynamoDB table to use On-Demand capacity mode is correct because the workload is highly unpredictable with long periods of complete inactivity, which eliminates idle capacity costs.

Step-by-Step Solution

Analyze the performance and storage requirements for the RDS for PostgreSQL database.

The database needs

200\text{ GB}

of storage and must handle up to

9,000\text{ IOPS}

during peak hours, with a baseline of

3,000\text{ IOPS}

Identifying the baseline and peak IOPS determines the storage type required.

Evaluate storage options for RDS PostgreSQL to find the most cost-effective tier.

gp3 storage allows provisioning

9,000\text{ IOPS}

independently of the storage capacity. gp2 would require scaling the volume to

3\text{ TB}

to get

9,000\text{ IOPS}

, which is highly wasteful, and io2 is much more expensive per GB and per IOPS.

gp3 decouples storage size from IOPS performance, minimizing costs for low storage/high IOPS workloads.

Analyze the access patterns and capacity requirements for the DynamoDB table.

The workload has highly unpredictable write bursts up to

2,000\text{ writes/sec}

followed by long periods of inactivity.

Understanding the variability of traffic determines the optimal DynamoDB capacity mode.

Compare DynamoDB On-Demand and Provisioned capacity modes.

On-Demand mode is ideal because the workload has unpredictable spikes and is completely idle for hours, meaning the company pays nothing during idle periods. Provisioned capacity with Auto Scaling is slower to respond to unpredictable spikes and incurs baseline costs during idle times.

On-Demand capacity mode optimizes costs by charging only for actual read/write requests without baseline idle charges.

Key Concept

Cost-Optimized Database Design and Capacity Planning

Question

Answer

Step-by-Step Solution

Key Concept

Common Mistakes

Topics