Welcome to Compute Labs Documentation

Welcome to the technical documentation for Compute Labs tools and services.

GPU Monitor & Validator

Our primary tool is the GPU Monitor & Validator - a professional-grade solution for monitoring and validating GPU performance in production environments.

Key Features

• Multi-vendor Support: Works with both NVIDIA (CUDA) and AMD (ROCm) GPUs
• Real-time Monitoring: Continuous performance tracking and health monitoring
• Hardware Validation: Verify GPU specifications against manufacturer specs
• Comprehensive Benchmarking: PyTorch-based stress tests and performance validation
• Anomaly Detection: Automated detection of performance issues and hardware problems

Latest Version: 3.1.2

The latest release includes:

• Fixed multi-GPU authentication issues
• Improved benchmark result accuracy
• Enhanced database architecture
• Better error handling for multi-GPU systems

Quick Links

Installation

Ubuntu/Debian Users

# Add PPA repository (Ubuntu) - non-interactive
sudo add-apt-repository ppa:cl-ax/gpu-validator -y
sudo apt update
sudo apt install cl-gpu-agent -y

Alternative Installation

# APT repository method
echo 'deb [trusted=yes] https://storage.googleapis.com/gpu_validator/apt-repo stable main' | \
  sudo tee /etc/apt/sources.list.d/computelabs-gpu-agent.list
sudo apt update
sudo apt install cl-gpu-agent

Getting Started

After installation, you can:

1. Validate your GPUs: cl-gpu-agent --mode validate
2. Run benchmarks: cl-gpu-agent --mode benchmark
3. Start monitoring service: sudo systemctl start cl-gpu-agent

Support

• Documentation: Read the full GPU Monitor documentation
• GitHub: gpu_validator_monitor
• Support Email: support@computelabs.ai
• Troubleshooting: Run sudo /opt/computelabs/gpu-validator/gpu-validator-troubleshoot