SAP-C02 · Question #606
A company needs to optimize the cost of its application on AWS. The application uses AWS Lambda functions and Amazon Elastic Container Service (Amazon ECS) containers that run on AWS Fargate. The…
The correct answer is D. Migrate the database to Aurora Serverless v1. Purchase Compute Savings Plans. For a write-heavy application with highly inconsistent load and long idle periods, migrating to Aurora Serverless for automatic scaling and cost-effectiveness, combined with Compute Savings Plans for Lambda and Fargate, is the most suitable solution.
Question
A company needs to optimize the cost of its application on AWS. The application uses AWS Lambda functions and Amazon Elastic Container Service (Amazon ECS) containers that run on AWS Fargate. The application is write-heavy and stores data in an Amazon Aurora MySQL database. The load on the application is not consistent. The application experiences long periods of no usage, followed by sudden and significant increases and decreases in traffic. The database runs on a memory optimized DB instance that cannot handle the load. A solutions architect must design a solution that can scale to handle the changes in traffic. Which solution will meet these requirements MOST cost-effectively?
Options
- AAdd additional read replicas to the database. Purchase Instance Savings Plans and RDS
- BMigrate the database to an Aurora DB cluster that has multiple writer instances. Purchase
- CMigrate the database to an Aurora global database. Purchase Compute Savings Plans and RDS
- DMigrate the database to Aurora Serverless v1. Purchase Compute Savings Plans.
How the community answered
(34 responses)- A3% (1)
- B12% (4)
- C21% (7)
- D65% (22)
Why each option
For a write-heavy application with highly inconsistent load and long idle periods, migrating to Aurora Serverless for automatic scaling and cost-effectiveness, combined with Compute Savings Plans for Lambda and Fargate, is the most suitable solution.
Adding read replicas only helps with read-heavy workloads, not the described write-heavy bottleneck, and Instance Savings Plans/RDS Reserved Instances do not provide the automatic scaling to zero needed for cost-effectiveness with intermittent loads.
Migrating to an Aurora DB cluster with multiple writer instances is not possible as Aurora only supports a single writer instance per cluster; this option misrepresents Aurora's architecture.
Aurora Global Database is for disaster recovery and low-latency global reads, not for addressing the primary issue of inconsistent load and cost-effective scaling for a single-region write-heavy database.
Migrating to Aurora Serverless v1 provides automatic scaling down to zero capacity during idle periods and scales up rapidly with demand, making it the most cost-effective solution for highly inconsistent workloads with long periods of no usage. Compute Savings Plans further reduce the cost of Lambda and Fargate usage.
Concept tested: Aurora Serverless, cost optimization, database scaling
Source: https://aws.amazon.com/rds/aurora/serverless/
Community Discussion
No community discussion yet for this question.