Objective 4.1 Deploy a Model AIP-210 Practice Questions (Page 5)

Questions 21–25

1. 21foundation · easy

   When deploying a model that processes personal data, what compliance consideration is important?

   • AEnsuring data privacy regulations are followed, such as GDPR
   • BMaximizing the model's training speed
   • CIncreasing the number of API endpoints
   • DReducing the model's file size
   Reveal Solution Discussion Select an answer first
2. 22expert · hard

   A team is deploying a model that requires a GPU for inference. They are using a container orchestration platform. The model is memory-intensive and sometimes fails with out-of-memory errors. What is the best way to handle this?

   • AUse a smaller model to reduce memory usage
   • BIgnore the errors because they are rare
   • CIncrease the memory limit for the container and add monitoring to detect memory usage
   • DReduce the batch size of inference requests
   Reveal Solution Discussion Select an answer first
4. 23expert · hard

   A startup has a limited budget and needs to deploy a model that serves predictions with a latency under 200 ms. The traffic is unpredictable, with occasional spikes. They are considering serverless functions and a small Kubernetes cluster. What is the most cost-effective approach that meets the latency requirement?

   • AA single VM running the model 24/7
   • BA Kubernetes cluster with a fixed number of nodes
   • CBatch processing with a nightly job
   • DServerless functions with a provisioned concurrency setting to reduce cold starts
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A team uses a model registry to manage model versions. They deploy a new version to production and later find a critical bug. They need to roll back quickly. What is the best practice to enable fast rollback?

   • ADelete the previous version from the registry to avoid confusion
   • BKeep the previous model version deployed in a separate environment and switch traffic back via a load balancer
   • CRely on the model registry to automatically roll back
   • DRetrain the model from scratch and deploy it
   Reveal Solution Discussion Select an answer first
6. 25foundation · easy

   An e-commerce website needs to display product recommendations instantly as users browse. Which deployment strategy is most appropriate?

   • AEdge deployment
   • BEmbedded deployment
   • CReal-time deployment
   • DBatch deployment
   Reveal Solution Discussion Select an answer first