Module gpu 

GPU Monitoring Module (MLOPS-005)

btop-inspired GPU monitoring for terminal training dashboard.

Toyota Way: Andon

Visual alerting system for immediate problem detection. Thermal throttling, memory pressure, and power limits trigger alerts.

Example

use entrenar::monitor::gpu::{GpuMonitor, GpuMetrics, GpuAlert};

let monitor = GpuMonitor::new()?;
let metrics = monitor.sample();
for m in &metrics {
    println!("GPU {}: {}°C, {}% util", m.device_id, m.temperature_celsius, m.utilization_percent);
}

Structs

AndonThresholds

Andon thresholds for GPU alerts

GpuAndonSystem

GPU Andon system for alert management

GpuMetrics

GPU metrics snapshot (inspired by btop’s GPU visualization)

GpuMetricsBuffer

GPU metrics history buffer (ring buffer)

GpuMonitor

GPU monitor that collects metrics

GpuProcess

Process using GPU resources

Enums

GpuAlert

GPU alert types for Andon system

Functions

format_gpu_panel

Format GPU metrics for terminal display

render_progress_bar

Render a progress bar for terminal display

render_sparkline

Render a sparkline from values