Network Appliance NetApp Certified AI Expert 認定 NS0-901 試験問題:

1. The architect needs to design an efficient data flow to move curated training sets from the central StorageGRID data lake to the high-performance NetApp ASA system used by the AI cluster. The process must be manageable from a single interface and should be automatable. Which two NetApp technologies should be used to implement this data pipeline stage? (Choose 2.)

A) NetApp SnapMirror to replicate data from the StorageGRID object store to the ASA's file system.
B) NetApp FabricPool to automatically tier the data from StorageGRID to the ASA.
C) A manual copy process using an S3 client on a bastion host.
D) A custom Python script using the NetApp DataOps Toolkit to trigger the data movement.
E) NetApp BlueXP copy and sync service to create and manage the data synchronization relationship between the S3 source and the NFS/CIFS destination.

2. The firm decides to implement a disaster recovery (DR) site for the "Advisor Assistant" application in a secondary data center. The Recovery Point Objective (RPO) is 15 minutes, and the Recovery Time Objective (RTO) is 4 hours. The design must protect both the document data lake and the vector database.
The primary site contains:
- Data Lake: NetApp StorageGRID
- Vector DB: NetApp AFF A-Series
Which combination of technologies and processes provides a complete and robust DR solution?
(Select all that apply.)

A) Use NetApp FlexCache at the DR site to cache data from the primary site.
B) Use BlueXP disaster recovery to orchestrate and automate the failover and failback workflows for the application and its data dependencies.
C) Rely on tape backups to be shipped to the DR site in the event of a disaster.
D) Use StorageGRID's built-in replication rules to replicate object data from the primary site's grid to a StorageGRID instance at the DR site.
E) Use the NetApp DataOps Toolkit to manually script the failover process.
F) Use NetApp SnapMirror to create an asynchronous replication relationship for the AFF A-Series volume containing the vector database, with a schedule of 10 minutes.

3. The firm decides to expand the "Advisor Assistant" project to a new team in a different department. This team needs its own isolated environment. An MLOps engineer attempts to submit a new GPU- intensive job for the new team, but it remains pending. The engineer checks the Run:AI scheduler logs and finds the following entry:
time="2025-07-11T16:45:00Z" level=info msg="Job ds-new-team-job1 cannot be scheduled.
Project 'new-team-project' has exceeded its GPU quota. Quota: 0, Requested: 1, Used: 0." What is the root cause of the scheduling failure?

A) The Kubernetes cluster has no available GPUs.
B) The Run:AI scheduler is offline and cannot process new jobs.
C) A Run:AI project quota has been configured for the new team, but it has been set to zero, effectively blocking them from using any GPU resources.
D) The job is requesting a specific type of GPU that is not available in the cluster.

4. A network administrator, attempting to harden the security of the data center, modifies a firewall access control list (ACL). Immediately afterward, the "Advisor Assistant" application pods can no longer mount their required NFS volumes from the AFF A-Series. The MLOps team confirms the pods are stuck in a 'ContainerCreating' state with a 'MountVolume.SetUp failed... connection timed out' error. The administrator provides the new, active firewall rule:
RULE | ACTION | PROTOCOL | SOURCE_IP_RANGE | DEST_IP_RANGE | DEST_PORT --|--|-
|--||--51 | ALLOW | TCP | 10.20.5.0/24 | 10.20.10.0/24 | 2049
What is the most likely reason for the mount failures?

A) The firewall rule is blocking the Portmapper/RPCbind service (TCP/UDP port 111), which is necessary for the initial NFS mount negotiation.
B) The firewall rule is blocking the NFS lock manager (NLM) protocol, which is required for file locking.
C) The 'SOURCE_IP_RANGE' is incorrect and does not include the Kubernetes pod IP addresses.
D) The firewall rule is blocking the NFSv4 protocol, which requires TCP port 2050.

5. The HPC cluster generates simulation data at an extremely high rate, requiring a storage system that can handle massively parallel writes from hundreds of compute nodes simultaneously. Which storage system and file protocol combination is the most appropriate choice for the HPC cluster's high-performance scratch space?

A) A NetApp ASA system serving a single, large NFS volume.
B) A NetApp StorageGRID system accessed via the S3 protocol.
C) A NetApp E-Series system serving a BeeGFS parallel file system.
D) A Cloud Volumes ONTAP instance with a standard file system.

質問と回答：