NVIDIA AI Infrastructure - NCP-AII무료 덤프문제 풀어보기
A company deploys a large language model training cluster containing multiple NVIDIA HGX servers. During benchmarking, GPUs within each server communicate efficiently, but gradient synchronization across servers introduces significantly higher latency than expected. The infrastructure engineer needs to identify the component primarily responsible for optimizing GPU- to-GPU communication between different servers while maintaining maximum bandwidth and minimum latency.
정답: B
설명: (Fast2test 회원만 볼 수 있음)
You are an infrastructure engineer tasked with validating a new AI training cluster before releasing it to users. Your team wants to perform a NeMo burn-in to ensure both hardware and software are reliable and ready for production workloads. Which of the following actions are required as part of a proper NeMo burn-in process? (Choose two.)
정답: B,C
설명: (Fast2test 회원만 볼 수 있음)
A Slurm-managed AI cluster contains both H100 and L40 GPU nodes. Several inference jobs requiring only modest GPU resources are repeatedly scheduled onto H100 nodes, delaying large distributed training jobs. Which scheduling strategy would best improve overall cluster efficiency?
정답: B
설명: (Fast2test 회원만 볼 수 있음)
A network architect deploys a Spectrum Ethernet fabric for a RoCE-based AI cluster. During stress testing, distributed training performance fluctuates significantly because packet loss occurs whenever multiple training jobs execute simultaneously. Which network feature should be prioritized to improve communication consistency?
정답: C
설명: (Fast2test 회원만 볼 수 있음)
If two ports must be connected, but one is SFP and one is QSFP, for example, to connect a 25 GbE HOST CHANNEL ADAPTER to a QSFP port capable of both 100 GbE and 25 GbE, which of the following solutions would best meet this requirement?
정답: C
설명: (Fast2test 회원만 볼 수 있음)
You are training a deep neural network using NCCL to coordinate communication across four GPUs in a single node. During early performance testing, you notice inconsistent scaling and longer-than-expected training times, even though all GPUs are being used. Which strategy would most effectively improve NCCL efficiency and collective operation performance in this setting?
정답: A
설명: (Fast2test 회원만 볼 수 있음)
A system administrator wants to enable vGPU virtualization on a DGX A100 system. What action should be taken first?
정답: A
설명: (Fast2test 회원만 볼 수 있음)
A user wants to restrict a Docker container to use only GPUs 0 and 2. Which command achieves this?
정답: A
설명: (Fast2test 회원만 볼 수 있음)
A system engineer needs to set the vGPU scheduling behavior for all GPUs to share the scheduling equally with the default time slice length. What command should be used?
정답: A
설명: (Fast2test 회원만 볼 수 있음)
You are tasked with validating the cooling system for a high-density AI cluster using NVIDIA Blackwell GPUs, which generate up to 120 kW of power per rack. What should you prioritize during validation?
정답: D
설명: (Fast2test 회원만 볼 수 있음)
When updating the firmware on an NVLink switch transceiver, how can an engineer apply new firmware without interrupting the network?
정답: D
설명: (Fast2test 회원만 볼 수 있음)
To validate bisectional bandwidth across two racks in a Spectrum-X Ethernet fabric, which NCCL test configuration isolates East-West traffic?
정답: B
설명: (Fast2test 회원만 볼 수 있음)
An engineer must ensure a BlueField-3 NIC firmware download matches the cluster's PSID.
Which step is critical before installation?
Which step is critical before installation?
정답: A
설명: (Fast2test 회원만 볼 수 있음)
What is the best practice for configuring memory in an NVIDIA certified server for optimizing performance?
정답: C
설명: (Fast2test 회원만 볼 수 있음)