Google Professional Machine Learning Engineer
Get started today
Ultimate access to all questions.
Comments
Loading comments...
Ultimate access to all questions.
A cybersecurity organization has developed a Dataflow pipeline using TensorFlow for anomaly detection. The pipeline ingests data via Pub/Sub and writes results to BigQuery. The organization aims to minimize system latency while ensuring scalability and cost-efficiency. Considering these constraints, which of the following steps should be taken to achieve the best performance? (Choose two options if E is available, otherwise choose one.)
A
Package the model prediction logic in Cloud Run, which is called by Dataflow. This approach leverages serverless computing for scalability but may introduce additional latency due to network calls.
B
Install the model to a Vertex AI endpoint, and invoke this endpoint in the Dataflow job. Vertex AI is optimized for low-latency predictions and offers managed services for ease of use.
C
Incorporate the model directly into the Dataflow job as a requirement, and utilize it for prediction. This method reduces external dependencies but may increase the complexity and processing time within the pipeline.
D
Install the model in a TFServing container on Google Kubernetes Engine, and invoke it in the Dataflow job. This option provides flexibility but requires additional management overhead and may not match Vertex AI's performance.
E
Combine options B and D by deploying the model to both Vertex AI and a TFServing container on GKE, then dynamically routing predictions based on current load to balance latency and cost.