NCP-AII Question 84
Select 3You are tasked with validating the operation of NVIDIA GPUs in a newly deployed AI training server. The server will run deep learning workloads using TensorFlow. Which of the following steps should you take to confirm the hardware is functioning correctly for the intended workloads?
- A
Run the NVIDIA System Management Interface (nvidia-smi) to check GPU availability and utilization.
- B
Execute a GPU diagnostics tool, such as DCGM (Data Center GPU Manager), to verify GPU health and performance.
- C
Install and test basic OpenGL applications to validate graphical rendering performance.
- D
Run a TensorFlow-based benchmark to ensure the GPUs are performing as expected under actual workload conditions.
- E
Check the BIOS settings to ensure Secure Boot is enabled for enhanced security.
Show answer and explanation
Correct answers: A, B, D
Explanation
Validating hardware operation for AI workloads involves ensuring that the GPUs are recognized, healthy, and capable of performing under realistic conditions. Tools like nvidia-smi and DCGM provide critical diagnostic information, while running a TensorFlow-based benchmark verifies that the hardware performs as expected for the intended workload. Checking unrelated features, such as OpenGL rendering or Secure Boot, does not contribute to validating GPU functionality for deep learning tasks.
- A. Correct.
nvidia-smi is a critical tool for checking GPU availability, health, and utilization. It helps ensure the hardware is properly detected and functional.
- B. Correct.
DCGM is specifically designed for monitoring and diagnosing GPU health and performance. Running diagnostics is a key step to validate hardware readiness.
- C. Incorrect.
Testing OpenGL applications is unrelated to AI training workloads, as OpenGL focuses on graphics rendering rather than AI computation.
- D. Correct.
Running a TensorFlow-based benchmark simulates actual workload conditions and confirms that the GPUs meet performance expectations for AI training.
- E. Incorrect.
While Secure Boot is a good practice for system security, it does not directly validate GPU hardware operation for AI workloads.