feat(monitoring): métriques conso électrique CPU+GPU (RAPL + ROCm power)

- node_exporter : activer --collector.rapl (CPU power package, storage-01 + gpu-01 + nœuds k8s)
- rocm_scraper : ajouter rocm_gpu_power_watts + rocm_gpu_power_cap_watts via sysfs hwmon
- gpu-01 : activer rocm_exporter_enabled
- kube-prometheus-stack : propager --collector.rapl aux DaemonSet Talos via prometheus-node-exporter.extraArgs

Déployé sur storage-01 et gpu-01. ArgoCD prend en charge les nœuds Talos.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
alkatrazz 2026-06-02 22:54:51 +02:00
parent 8236448859
commit 91e648f55e
4 changed files with 19 additions and 1 deletions

View file

@ -6,6 +6,9 @@ llama_model_path: "/mnt/models/bartowski/Qwen3-8B-GGUF/Qwen3-8B-Q4_K_M.gguf"
llama_model_alias: "qwen3-8b"
llama_ctx_size: 32768
# ROCm scraper — métriques GPU dans Prometheus
rocm_exporter_enabled: true
# Montages NFS (surcharge les defaults du rôle nfs_client)
nfs_client_mounts:
- local_path: /mnt/nfs