Stress Test the GPU Nodes POWEREDGE-XE-OPERATE Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   Which command would you use to run a GPU stress test that loads the GPU to maximum capacity and reports any errors?

   • Agpu-burn
   • Bstress --cpu 8
   • Cnvidia-smi dmon
   • Dnvidia-smi -q
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Before starting a GPU stress test, which command should you run to verify that the NVIDIA driver is loaded and the GPUs are visible to the system?

   • Admesg | grep -i error
   • Btop
   • Cnvidia-smi
   • Dgpu-burn
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   You need to stress test a single GPU on a PowerEdge XE server that has multiple GPUs. You want to isolate the test to that specific GPU to avoid affecting other workloads. Which command should you use?

   • A`nvidia-smi -i 0 -c 0`
   • B`nvidia-smi dmon -i 0`
   • C`CUDA_VISIBLE_DEVICES=0 gpu-burn 600`
   • D`gpu-burn 600`
   Reveal Solution Discussion Select an answer first
5. 9expert · hard

   A stress test on a PowerEdge XE server produced the following output: `gpu-burn` completed with '0 errors', but `nvidia-smi dmon` showed that GPU utilization never exceeded 60% and power draw stayed at 50% of the maximum. What is the most likely cause?

   • AThe stress test did not fully load the GPU because the workload was not intensive enough.
   • BThe power supply is insufficient, limiting the GPU's power draw.
   • CThe GPU is faulty and cannot reach full utilization.
   • DThe driver is not properly installed, causing reduced performance.
   Reveal Solution Discussion Select an answer first
6. 10foundation · easy

   Which command provides a real-time view of GPU utilization, temperature, and power consumption during a stress test?

   • Atop
   • Bnvidia-smi dmon
   • Cgpu-burn
   • Dnvidia-smi -q
   Reveal Solution Discussion Select an answer first