Google Professional Data Engineer
Get started today
Ultimate access to all questions.
Ultimate access to all questions.
As the manager of a system that processes messages from IoT devices worldwide, you're facing challenges due to unpredictable spikes in message volume, especially from batched messages, making the current single on-premises Kafka cluster in the us-east region difficult and costly to manage. What Google Cloud architecture would best address this issue?
Explanation:
The optimal solution involves connecting an IoT gateway to Cloud Pub/Sub and utilizing Cloud Dataflow for processing messages (Option C). This approach efficiently manages the sporadic spikes in message volume by leveraging Cloud Pub/Sub's scalability for incoming loads and Cloud Dataflow's managed processing capabilities. Option A, which suggests virtualizing a Kafka cluster and using Cloud Load Balancing, may not adequately address scalability and cost issues. Option B, involving Edge TPUs for data storage and transmission, is less scalable for long-term needs. Option D, integrating Cloud Dataflow with the Kafka cluster, might not resolve the underlying issue of volume spikes, as the Kafka cluster could still struggle with high message volumes.