Answer-first summary for fast verification
Answer: Switch to a Cloud Pub/Sub topic for data ingestion instead of a REST API endpoint, allowing the ingestion application to consume messages from the topic.
Opting to write data to a Cloud Pub/Sub topic rather than a REST API endpoint is the correct choice. Cloud Pub/Sub efficiently buffers data during spikes, ensuring no data loss. While Cloud Storage is ideal for large objects like images or videos, Cloud Pub/Sub is more suited for small, frequent data such as telemetry from IoT sensors. Cloud SQL lacks the necessary scalability and low latency for this scenario, and HBase on Hadoop introduces unnecessary administrative complexity and scalability issues compared to Cloud Pub/Sub. For more details, refer to [Google Cloud documentation](https://cloud.google.com/dataflow/docs/concepts/streaming-with-cloud-pubsub).
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
A manufacturer of delivery drones is enhancing their data analysis pipeline to preemptively detect part failures. The drones are equipped with various sensors that transmit performance and environmental data to an analytics pipeline. Currently, this data is sent to a REST API endpoint, which sometimes cannot handle the influx, leading to data loss. To mitigate this, machine learning engineers have requested a modification to the ingestion process. What would be your recommended solution?
A
Utilize a Cloud Storage bucket for data ingestion instead of a REST API endpoint, enabling the ingestion application to read from the bucket.
B
Switch to a Cloud Pub/Sub topic for data ingestion instead of a REST API endpoint, allowing the ingestion application to consume messages from the topic.
C
Implement a Cloud SQL Postgres database for data ingestion, with the ingestion application querying the database directly.
D
Deploy a Hadoop cluster on Compute Engine using managed instance groups, storing data in an HBase database for the application to query.