Module monitor 

GPU Monitoring, Tracing, and Visualization (TRUENO-SPEC-010)

This module provides comprehensive GPU monitoring capabilities:

• Device discovery and information
• Real-time metrics collection
• Cross-platform support (wgpu + optional CUDA)

Design Philosophy

Dual-Backend Architecture: Supports both wgpu (cross-platform) and trueno-gpu (native CUDA) for maximum flexibility.

References

• Nickolls et al. (2008): GPU parallel computing model
• Gregg (2016): Flame graph visualization
• btop: NVML/ROCm SMI patterns

Example

use trueno::monitor::{GpuDeviceInfo, GpuMonitor};

// Query device information
let info = GpuDeviceInfo::query()?;
println!("GPU: {}", info.name);
println!("VRAM: {} MB", info.vram_total / (1024 * 1024));

// Start background monitoring
let monitor = GpuMonitor::new(0)?;
let metrics = monitor.collect()?;
println!("VRAM used: {} / {}", metrics.memory.used, metrics.memory.total);

Re-exports

pub use backends::cuda_monitor_available;

pub use backends::enumerate_cuda_devices;

pub use backends::query_cuda_device_info;

pub use backends::query_cuda_memory;

Modules

backends

wgpu and CUDA backend implementations for GPU monitoring

Structs

DeviceId

Device identification

GpuClockMetrics

GPU clock metrics

GpuDeviceInfo

GPU device information (TRUENO-SPEC-010)

GpuMemoryMetrics

GPU memory metrics

GpuMetrics

Complete GPU metrics snapshot

GpuMonitor

GPU Monitor for real-time metrics collection (TRUENO-SPEC-010)

GpuPcieMetrics

GPU PCIe metrics

GpuPowerMetrics

GPU power metrics

GpuThermalMetrics

GPU thermal metrics

GpuUtilization

GPU utilization metrics

MonitorConfig

Configuration for GPU monitoring

Enums

DeviceType

GpuBackend

GPU compute backend

GpuVendor

GPU vendor identifier based on PCI vendor ID

MonitorError

Errors from GPU monitoring operations

Traits

ComputeDevice

Unified compute device abstraction (TRUENO-SPEC-020)