CertyRush
Đang tải...
C CertyRush
Câu hỏi free preview

Câu 6: NCP-AIO: NCP - AI Operations

You are managing a Slurm cluster with multiple GPU nodes, each equipped with different types of GPUs. Some jobs are being allocated GPUs that should be reserved for other purposes, such as display rendering. How would you ensure that only the intended GPUs are allocated to jobs?

Nội dung câu hỏi

You are managing a Slurm cluster with multiple GPU nodes, each equipped with different types of GPUs. Some jobs are being allocated GPUs that should be reserved for other purposes, such as display rendering. How would you ensure that only the intended GPUs are allocated to jobs?

Các lựa chọn

Đáp án được giữ gọn theo nhãn A, B, C, D trong phần bình chọn tương tác.

  1. A. Verify that the GPUs are correctly listed in both gres.conf and slurm.conf, and ensure that unconfigured GPUs are excluded. — đáp án hiện tại
  2. B. Use nvidia-smi to manually assign GPUs to each job before submission.
  3. C. Reinstall the NVIDIA drivers to ensure proper GPU detection by Slurm.
  4. D. Increase the number of GPUs requested in the job script to avoid using unconfigured GPUs.

Cộng đồng

0 bình luận công khai. Tên thành viên được ẩn một phần.

Chưa có bình luận. Mở giao diện tương tác để bắt đầu thảo luận.

Câu hỏi liền kề