nerdexam
Google

PROFESSIONAL-DATA-ENGINEER · Question #268

A TensorFlow machine learning model on Compute Engine virtual machines (n2-standard -32) takes two days to complete framing. The model has custom TensorFlow operations that must run partially on a CPU

The correct answer is A. Use Cloud Composer to subscribe to a Pub/Sub tope and can the Python API.. Explanation/Reference: Explanation/Reference: https://cloud.google.com/functions/docs/calling/pubsub#deployment

Submitted by carlos_mx· Mar 30, 2026Operationalizing machine learning models

Question

A TensorFlow machine learning model on Compute Engine virtual machines (n2-standard -32) takes two days to complete framing. The model has custom TensorFlow operations that must run partially on a CPU You want to reduce the training time in a cost-effective manner. What should you do?

Options

  • AUse Cloud Composer to subscribe to a Pub/Sub tope and can the Python API.
  • BUse a Pub/Sub push subscription to trigger a Cloud Function to pass the data to tie Python API.
  • CWrite an application that makes a queue in a NoSQL database
  • DWrite an application hosted on a Compute Engine instance that makes a push subscription to the Pub/Sub topic

How the community answered

(58 responses)
  • A
    71% (41)
  • B
    9% (5)
  • C
    17% (10)
  • D
    3% (2)

Explanation

Explanation/Reference: Explanation/Reference: https://cloud.google.com/functions/docs/calling/pubsub#deployment

Topics

#TensorFlow#custom ops#CPU training#compute optimization

Community Discussion

No community discussion yet for this question.

Full PROFESSIONAL-DATA-ENGINEER Practice