DevOps and MLOps DY0-001 Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   In continuous integration (CI), what is the primary purpose of automatically building and testing every code change as soon as it is committed to the shared repository?

   • ATo automatically deploy every successful build directly to production without any human approval
   • BTo replace the need for any manual testing or quality assurance activities in the software development lifecycle
   • CTo ensure that only the most experienced developers can commit code to the shared repository
   • DTo ensure that the integrated codebase remains in a working state and that integration issues are detected early
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which capability is provided by container orchestration tools?

   • AElimination of the need for load balancing in the application
   • BStatic allocation of containers to specific servers that never changes
   • CManual restart of containers only when a developer requests it
   • DAutomatic scaling of container instances based on demand
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   An e-commerce company runs its recommendation engine as a set of microservices in Kubernetes. During a flash sale, the number of requests to the recommendation service spikes dramatically. The service is stateless and can be scaled horizontally. Which Kubernetes feature should the team configure to handle this automatically?

   • AHorizontal Pod Autoscaler (HPA) based on CPU utilization or custom metrics like request rate.
   • BA Vertical Pod Autoscaler (VPA) that increases the CPU and memory limits of the existing pods.
   • CA NetworkPolicy that prioritizes traffic from the flash sale users.
   • DA Kubernetes CronJob that scales the deployment to a fixed higher number of replicas during the sale period.
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   An online retailer uses a CI/CD pipeline to deploy its product-recommendation model. The pipeline is configured to automatically deploy the model to production whenever a new version passes all automated tests. The team wants to add a safety mechanism to catch issues that automated tests might miss, such as a subtle drop in recommendation quality. Which addition to the pipeline would best serve this purpose?

   • AAdd more unit tests to the CI pipeline to increase code coverage.
   • BDeploy the model to a staging environment and run a full regression test suite.
   • CRequire a manual approval step before the model is deployed to production.
   • DImplement a canary deployment step that routes a small percentage of traffic to the new model and compares its performance to the old model before a full rollout.
   Reveal Solution Discussion Select an answer first
6. 10expert · hard

   A large enterprise runs a critical ML inference service on a Kubernetes cluster. The service must meet a strict 99.99% availability SLA. The team is evaluating different deployment strategies to minimize downtime during model updates. They also need to ensure that a failed deployment can be rolled back quickly. Which strategy best balances these requirements?

   • ABlue/green deployment with automated health checks and instant traffic switchover.
   • BRecreate deployment, where all old pods are terminated before new ones are created.
   • CCanary deployment with a 1% traffic split and manual monitoring for 24 hours.
   • DRolling deployment with a large max surge and max unavailable setting.
   Reveal Solution Discussion Select an answer first