GPU Monitor

A real-time GPU and System Memory monitoring tool with color-coded visual indicators.

Project Structure

GPU_monitor/
├── src/                # Source code
│   └── monitor_gpus.py # Main application file
├── scripts/            # Utility and test scripts
│   ├── run_gpu_monitor.bat
│   ├── run_tests.bat
│   ├── test_monitor_gpus.py
│   └── toggle_gpu.bat
├── docs/              # Documentation
├── logs/              # Application and test logs
├── monitor_gpus/      # Virtual environment
├── .gitignore        # Git ignore file
└── requirements.txt   # Project dependencies

For detailed project architecture, see ARCHITECTURE.md.

Features

• Real-time monitoring of GPU metrics:
  • Utilization
  • Memory usage
  • Temperature
  • Power consumption
  • PCIe information
• System RAM monitoring
• Color-coded indicators for quick status assessment:
  • Blue: Low utilization/temperature
  • Green: Normal operation
  • Yellow: Warning level
  • Red: Critical level
• Configurable refresh rates
• Support for multiple GPUs
• Automated test suite with coverage reporting
• Comprehensive logging with timestamps and process IDs

Requirements

• Python 3.6+
• NVIDIA GPU(s)
• NVIDIA drivers installed
• Windows OS (batch scripts are Windows-specific)

Quick Start

1. Clone the repository:

git clone https://github.com/Xza85hrf/gpu-monitor.git
cd gpu-monitor

2. Run the setup script:

scripts/run_gpu_monitor.bat

This will:

• Create a virtual environment
• Install required packages
• Start the GPU monitor

Manual Installation

1. Create and activate virtual environment:

python -m venv monitor_gpus
monitor_gpus\Scripts\activate

2. Install required packages:

pip install -r requirements.txt

3. Run the monitor:

python src/monitor_gpus.py

Usage

Basic usage:

python src/monitor_gpus.py

With custom options:

# Set custom update interval (in seconds)
python src/monitor_gpus.py --interval 1.0

# Monitor specific GPUs by index
python src/monitor_gpus.py --gpu-indices 0 1

# Set custom refresh rate
python src/monitor_gpus.py --refresh-rate 0.5

Color Indicators

Temperature

• Blue: < 40°C (Cool/Ideal)
• Green: 40-60°C (Normal)
• Yellow: 60-80°C (Warning)
• Red: > 80°C (Critical)

Utilization

• Blue: < 30% (Light load)
• Green: 30-70% (Normal)
• Yellow: 70-90% (Heavy)
• Red: > 90% (Critical)

Memory Usage

• Blue: < 30% (Low usage)
• Green: 30-70% (Normal)
• Yellow: 70-90% (High)
• Red: > 90% (Critical)

Development

Running Tests

scripts/run_tests.bat

This will:

• Run all tests with verbose output
• Generate a coverage report
• Store test logs in logs/test_gpu_monitor.log

Log Files

Application logs are stored in logs/gpu_monitor.log with the following information:

• Timestamp
• Process ID
• Log level
• Message

Contributing

1. Fork the repository
2. Create your feature branch
3. Commit your changes
4. Create a pull request

For detailed contribution guidelines, see CONTRIBUTING.md.

License

MIT License

Support

For issues and feature requests, please use the GitHub issue tracker.