NetApp Certified AI Expert Exam: NS0-901 Exam Premium File as PDF & Test Engine

The company's compliance officer mandates that all raw simulation data in the StorageGRID data lake must be protected from modification or deletion for a period of seven years. Which two StorageGRID features should be implemented to meet this strict, long-term data immutability requirement? (Choose 2.)

• A. Use NetApp SnapMirror to replicate the data to a second StorageGRID system.
• B. Configure an Information Lifecycle Management (ILM) rule that applies to the bucket and sets a retention period of seven years.
• C. Encrypt all objects in the bucket using server-side encryption (SSE).
• D. Create a new S3 bucket with Object Lock enabled in Compliance mode.
• E. Enable Autonomous Ransomware Protection (ARP) on the StorageGRID system.

An architect is designing a data pipeline for a predictive AI model that will forecast retail sales.
The pipeline must be robust, version-controlled, and efficient.
The proposed data flow is as follows:
1. Ingest: Raw sales data is copied daily from multiple point-of-sale (POS) systems to a central staging area on an on-premises ONTAP cluster.
2. Prepare: The raw data is messy. A data engineering team needs a clean, isolated, and writable copy of the latest daily data to perform cleansing and feature engineering tasks without impacting the original raw data.
3. Train: Once prepared, the cleansed dataset is used to retrain the predictive model on a GPU cluster.
This step must be repeatable with the exact same dataset for compliance.
4. Deploy: The newly trained model is pushed to production inference servers.
Which combination of NetApp technologies best supports this entire predictive AI lifecycle?
(Select all
that apply.)

• A. Use NetApp StorageGRID as the primary storage for the high-performance training stage.
• B. Use NetApp XCP to efficiently aggregate the raw sales data from POS systems into the central staging area.
• C. Use BlueXP backup and recovery to perform the initial data ingest from the POS systems.
• D. Use NetApp Snapshots on the prepared dataset volume just before training to create an immutable, point-in-time version for compliance and reproducibility.
• E. Use a RAG architecture for the sales forecasting model.
• F. Use NetApp FlexClone to create an instantaneous, space-efficient, writable copy of the daily raw data for the data preparation stage.

A research institute is designing an infrastructure to support its entire AI drug discovery pipeline.
The pipeline has two distinct workload requirements:
1. Training: A team of data scientists needs to train several large transformer models simultaneously using a 500 TB dataset of genomic sequences. This process requires maximum data throughput to keep the GPUs saturated.
2. Inference: Once trained, the models are deployed to an internal web portal where researchers submit individual protein sequences for analysis. These queries must return results with the lowest possible latency.
Which infrastructure design best satisfies both requirements? (Choose 2.)

• A. Deploy a large NetApp ASA cluster with GPUDirect Storage enabled for the training environment.
• B. Configure QoS minimums on the training volumes to ensure they do not impact inference performance.
• C. Implement NetApp FlexCache on smaller nodes at the network edge to serve the inference requests.
• D. Use a single, large Cloud Volumes ONTAP instance in a public cloud to handle both workloads to simplify management.
• E. Use NetApp StorageGRID as the primary storage for both the training and low-latency inference workloads.

An MLOps engineer is troubleshooting a failed Kubeflow pipeline step. The step was designed to create a clone of a dataset for a training job using the NetApp DataOps Toolkit. The pod logs for the failed pipeline step show the following:
Traceback (most recent call last):
File "create_clone.py", line 15, in <module>
clone_pvc(source_pvc_name="training-data-v2", new_pvc_name="train-job-34a-data") NameError: name 'clone_pvc' is not defined The engineer reviews the Python script for the pipeline step:
# create_clone.py
import os
from netapp_dataops.k8s import create_pvc
# Other code
print("Cloning source dataset for training run...")
clone_pvc(
source_pvc_name="training-data-v2",
new_pvc_name="train-job-34a-data"
)
print("Clone created successfully.")
What is the cause of the error?

• A. The NetApp DataOps Toolkit is not installed in the container image used for this pipeline step.
• B. The Python script is attempting to use the 'clone_pvc' function, but it was not imported from the
  'netapp_dataops.k8s' library.
• C. The source PVC 'training-data-v2' does not exist.
• D. The Kubernetes cluster is not running NetApp Trident.