config.yaml

# MyGPU Configuration

cluster:
  name: "My GPU Cluster"
  name: "Prototype Cluster Monitor"
  nodes: []  # Add nodes for multi-node monitoring
  # Example for future multi-node setup:
  # - hostname: gpu-server-01
  #   ssh_user: admin
  # - hostname: gpu-server-02
  #   ssh_user: admin

monitoring:
  interval_seconds: 5
  history_retention_hours: 168  # 1 week to change

alerts:
  # GPU alerts
  gpu_temperature_warn: 80      # Warning at 80°C
  gpu_temperature_critical: 90  # Critical at 90°C
  gpu_memory_usage_warn: 90     # Warning at 90% memory
  gpu_utilization_low: 10       # Alert if GPU idle
  
  # System alerts (percentages)
  cpu_usage_warn: 90
  memory_usage_warn: 90
  disk_usage_warn: 90

web:
  host: "0.0.0.0"
  port: 8090
  
storage:
  type: sqlite
  path: ./metrics.db