Refine burn UI and NVIDIA stress flows

This commit is contained in:
2026-04-05 13:43:43 +03:00
parent 25af2df23a
commit 38e79143eb
18 changed files with 825 additions and 229 deletions

View File

@@ -1,6 +1,10 @@
# NVIDIA DCGM (Data Center GPU Manager) — dcgmi diag for acceptance testing.
# DCGM 4 is packaged per CUDA major. The image ships NVIDIA driver 590 with CUDA 13 userspace,
# so install the CUDA 13 build plus proprietary diagnostic components explicitly.
# NVIDIA DCGM (Data Center GPU Manager).
# Validate uses dcgmi diagnostics; Burn uses dcgmproftester as the official
# NVIDIA max-compute recipe. The smoketest/runtime contract treats
# dcgmproftester as required in the LiveCD.
# DCGM 4 is packaged per CUDA major. The image ships NVIDIA driver 590 with
# CUDA 13 userspace, so install the CUDA 13 build plus proprietary components
# explicitly.
datacenter-gpu-manager-4-cuda13=1:%%DCGM_VERSION%%
datacenter-gpu-manager-4-proprietary=1:%%DCGM_VERSION%%
datacenter-gpu-manager-4-proprietary-cuda13=1:%%DCGM_VERSION%%