MLS-C01 Question 46
Select 2A company is using Amazon Managed Service for Apache Flink to process streaming data from IoT devices in near real-time. The team needs to ensure that the application can handle fluctuations in the data stream volume by dynamically scaling the resources, while also optimizing costs. Which of the following features or configurations should they use?
- A
Enable Auto Scaling for Kinesis Data Streams as the input source.
- B
Use Apache Flink's Checkpointing feature to persist application state.
- C
Configure parallelism in the Flink application to match the expected peak throughput.
- D
Enable Amazon Managed Service for Apache Flink's auto-scaling feature.
- E
Use Flink's Savepoints to manually scale the application when needed.
Show answer and explanation
Correct answers: A, D
Explanation
To handle fluctuations in data stream volume and optimize costs, enabling auto-scaling for both the input source (e.g., Kinesis Data Streams) and the Flink application itself is essential. These features allow the application to dynamically scale resources up or down based on traffic, ensuring efficiency and cost-effectiveness. Checkpointing and Savepoints are useful for state management but do not directly address the scaling requirements in this scenario. Configuring parallelism statically does not provide the needed flexibility to handle changing workloads.
- A. Correct.
Enabling Auto Scaling for Kinesis Data Streams ensures that the input source can dynamically adjust to handle fluctuating data volumes, preventing bottlenecks.
- B. Incorrect.
Checkpointing is used for state persistence and fault tolerance, but it does not directly address dynamic scaling or cost optimization.
- C. Incorrect.
Manually configuring parallelism may help with performance, but it does not dynamically adjust resources based on the volume of data or optimize costs.
- D. Correct.
Amazon Managed Service for Apache Flink's auto-scaling feature automatically adjusts the resources allocated to the application, optimizing performance and cost.
- E. Incorrect.
Savepoints are used for manually pausing and resuming stateful Flink applications during updates or migrations, but they do not provide dynamic scaling.