NVIDIA-Certified Professional: AI Infrastructure Practice Test
Gana confianza para NVIDIA-Certified Professional: AI Infrastructure. Practica los conceptos, comprende las respuestas y refuerza tus conocimientos pregunta a pregunta.
Probar una preguntaDescripción y detalles del examen
El examen NVIDIA-Certified Professional: AI Infrastructure valida las habilidades necesarias para diseñar, implementar y gestionar soluciones de infraestructura de IA basadas en las plataformas de computación acelerada de NVIDIA. Este examen evalúa la capacidad del candidato para arquitectar clústeres acelerados por GPU, configurar redes y almacenamiento para entrenamiento distribuido, optimizar tuberías de inferencia y solucionar cuellos de botella de rendimiento en entornos de IA en producción. Cubre temas como el stack de software NVIDIA AI Enterprise, configuraciones multi-GPU y multi-nodo, contenedorización con NVIDIA NGC y herramientas de monitoreo como DCGM y NVSM.
Preguntas de Muestra
Elige una respuesta y consulta la explicación para ver cómo funciona la práctica.
A node fails the "Confirm FW/SW on transceivers" check because several 400 Gb/s transceivers report a firmware version one minor release behind the approved BOM. All links train at full speed and pass ibdiagnet with zero errors. Under Cluster Test and Verification, what is the correct decision?
An administrator has just completed OS installation on the two head nodes of a new Base Command Manager cluster. The next requirement is to configure High Availability so that the secondary head can take over if the primary fails, using the standard BCM HA mechanism with shared storage for the cluster database.
The final Cluster Test and Verification phase requires running burn-in workloads that exercise the full software stack. Which three tests together satisfy the explicit requirements to "Perform NCCL burn-in", "Perform HPL burn-in", and "Perform NeMo(TM) burn-in"?
When performing initial configuration of a DGX H100, the engineer must also handle TPM. Which statement correctly places TPM configuration in the System and Server Bring-up sequence?
An infrastructure engineer has just completed physical racking and NDR InfiniBand cabling of the first eight HGX H100 nodes in a new AI factory rack. The PDUs are powered and the out-of-band management network is reachable from the laptop. According to the documented System and Server Bring-up sequence, what is the immediate next required action before any firmware or OS-level changes?
Qué temas cubre este examen
Usa las ponderaciones publicadas de los dominios para planificar tu estudio. Los resultados de práctica no predicen tu puntuación en el examen de certificación.