SAP-C02 Question 328
Select 2A company is running a real-time analytics application on AWS. The application processes large volumes of data received from IoT devices globally. The company uses Amazon Kinesis Data Streams to ingest the data, AWS Lambda for processing, and Amazon DynamoDB for storage. Recently, the company has observed increased latency in Lambda processing due to a sudden burst of incoming data. As a Solutions Architect, what solution would you recommend to handle this burst effectively while maintaining real-time processing?
- A
Enable Kinesis Enhanced Fan-Out to increase the throughput for Lambda consumers.
- B
Increase the shard count of the Kinesis Data Stream to handle higher data throughput.
- C
Configure DynamoDB Auto Scaling to dynamically adjust the read and write capacity.
- D
Use AWS Lambda Reserved Concurrency to ensure enough resources are available for processing.
- E
Switch from Kinesis Data Streams to Amazon SQS for data ingestion to reduce latency.
Show answer and explanation
Correct answers: A, B
Explanation
To handle the burst of incoming data effectively, the solution must address the increased throughput requirements at the Kinesis Data Stream level. Enabling Enhanced Fan-Out allows Lambda consumers to process data independently and at higher throughput, while increasing the shard count ensures the Kinesis Data Stream can handle the surge in data. These two options address the root cause of the issue without requiring a major architectural overhaul.
- A. Correct.
Enabling Kinesis Enhanced Fan-Out allows multiple consumers, including Lambda, to receive data independently at a higher throughput. This is effective in reducing bottlenecks during bursts.
- B. Correct.
Increasing the shard count in Kinesis Data Streams increases the stream's capacity to handle more data, which is critical during a sudden burst.
- C. Incorrect.
DynamoDB Auto Scaling adjusts capacity for DynamoDB tables but does not address the issue with Lambda processing latency caused by a sudden data burst in Kinesis.
- D. Incorrect.
Using Lambda Reserved Concurrency ensures a specific number of concurrent invocations, but this does not address the root cause of increased latency due to the burst in Kinesis data ingestion.
- E. Incorrect.
Switching to Amazon SQS might reduce latency in certain cases, but it would require a significant architectural change and would not maintain the real-time capabilities of Kinesis Data Streams.