nerdexam
Google

PROFESSIONAL-MACHINE-LEARNING-ENGINEER · Question #337

You work at an organization that manages a popular payment app. You built a fraudulent transaction detection model by using scikit-learn and deployed it to a Vertex AI endpoint. The endpoint is…

The correct answer is D. Increase the number of maximum replicas to 6 nodes, each with 1 machine. Increasing the number of replicas allows the endpoint to handle fluctuating traffic by distributing requests across multiple instances of the same machine type. This horizontal scaling is more cost-effective than upgrading to a much larger machine or using specialized…

Submitted by kwame.gh· Apr 18, 2026Monitoring, optimizing, and maintaining ML solutions

Question

You work at an organization that manages a popular payment app. You built a fraudulent transaction detection model by using scikit-learn and deployed it to a Vertex AI endpoint. The endpoint is currently using 1 with 2 vCPUs and 8 GB of memory. e2-standard-2 machine You discover that traffic on the gateway fluctuates to four times more than the endpoint's capacity. You need to address this issue by using the most cost-effective approach. What should you do?

Options

  • ARe-deploy the model with a TPU accelerator.
  • BChange the machine type to e2-highcpu-32 with 32 vCPUs and 32 GB of memory.
  • CSet up a monitoring job and an alert for CPU usage. If you receive an alert, scale the vCPUs as
  • DIncrease the number of maximum replicas to 6 nodes, each with 1 machine.

How the community answered

(26 responses)
  • A
    4% (1)
  • B
    8% (2)
  • C
    12% (3)
  • D
    77% (20)

Explanation

Increasing the number of replicas allows the endpoint to handle fluctuating traffic by distributing requests across multiple instances of the same machine type. This horizontal scaling is more cost-effective than upgrading to a much larger machine or using specialized accelerators when the workload is CPU-bound and fluctuating.

Topics

#Vertex AI Deployment#Auto-scaling#Cost Optimization#Model Serving Performance

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full PROFESSIONAL-MACHINE-LEARNING-ENGINEER Practice