1Z0-1110-25 OCI 2025 Data Science Professional Practice Questions
Prepare for 1Z0-1110-25 with more than an answer.
- 1
You are implementing a CI/CD pipeline for machine learning. You need to verify that a model artifact is functional before deploying it to production. Which OCI Data Science feature allows you to run a mock deployment test on the artifact?
Show answer details
Correct answer: A
Model Introspection in OCI Data Science allows you to run a series of tests on the model artifact (such as checking
score.pysyntax and dependency loading) to ensure it is valid and deployable before creating an actual Model Deployment. - 2
Case Study:
A retail company is migrating its on-premises demand forecasting system to OCI. The system consists of:
- Daily data ingestion from Oracle Database.
- A retraining process that runs every Sunday.
- A batch inference process that runs every night.
They require a solution that orchestrates these steps, manages dependencies, and provides failure alerts. The team wants to use Python for defining the workflow.
Which OCI Data Science architecture best fits these requirements?
Show answer details
Correct answer: A
OCI Data Science Pipelines is the native orchestration service. It allows defining dependencies between steps (Ingestion -> Training -> Inference) using Python/YAML. It integrates with OCI Events for scheduling and alerting, meeting all requirements.
- 3
You are configuring a Model Deployment. You expect the traffic to be variable, with spikes during business hours. You want to ensure the deployment scales automatically based on load. Which metric is primarily used by the OCI Model Deployment auto-scaling policy?
Show answer details
Correct answer: C
OCI Model Deployment auto-scaling policies are defined based on the throughput (number of requests) the load balancer receives. You define thresholds for scaling out (adding instances) or scaling in.
- 4
Which OCI IAM policy statement allows a Data Science user group
DS-Usersto manage their own model artifacts in the Model Catalog within a specific compartmentProject-A?Show answer details
Correct answer: A
The resource type for models in the catalog is
data-science-models. The verbmanageprovides full permissions (create, delete, update). - 5
You are architecting a secure OCI Data Science environment for a financial institution. The compliance team requires that all notebook sessions must be deployed in a private subnet and must not have direct internet access. However, data scientists need to access public PyPI repositories to install specific libraries. Which networking configuration provides the most secure and compliant solution?
Show answer details
Correct answer: C
While a Service Gateway handles OCI internal traffic (like Object Storage), access to external repositories like PyPI requires a NAT Gateway. To meet strict compliance, combining the NAT Gateway with Network Security Groups (NSGs) to whitelist specific outbound traffic is the optimal secure approach, as opposed to a completely open NAT or trying to mirror the entire PyPI repo which is operationally complex.
- 6
A data scientist is using the Oracle Accelerated Data Science (ADS) SDK to prepare a dataset for training. The dataset contains several columns with sensitive PII that need to be redacted before model training. Which ADS feature should be used to automatically identify and treat these sensitive columns?
Show answer details
Correct answer: D
The ADS SDK includes built-in data profiling capabilities that can detect PII (Personally Identifiable Information) when loading datasets using the
ads.dataset.factory. It provides recommendations and transformations to mask or drop these columns.
