kaaskoek232-ComfyUI-MemoryManagement/DEPLOYMENT.md

ComfyUI Memory Management - Deployment Guide

🚀 Production Deployment Instructions

This guide provides step-by-step instructions for deploying the ComfyUI Memory Management custom nodes in production environments.

📋 Prerequisites

System Requirements

• Python: 3.8 or higher
• PyTorch: 1.9.0 or higher
• ComfyUI: Latest stable version
• RAM: Minimum 8GB (16GB+ recommended)
• GPU: NVIDIA GPU with CUDA support (optional but recommended)

Required Dependencies

• psutil>=5.9.0 - System monitoring
• torch>=1.9.0 - GPU operations

🔧 Installation Methods

Method 1: ComfyUI Manager (Recommended)

1. Install ComfyUI Manager (if not already installed):

   cd ComfyUI/custom_nodes
   git clone https://github.com/ltdrdata/ComfyUI-Manager.git
   

2. Use Manager to Install:

   • Launch ComfyUI
   • Open ComfyUI Manager
   • Search for "Memory Management"
   • Click Install
   • Restart ComfyUI

Method 2: Manual Git Installation

1. Navigate to ComfyUI custom nodes directory:

   cd /path/to/ComfyUI/custom_nodes
   

2. Clone this repository:

   git clone https://github.com/ComfyUI/ComfyUI-MemoryManagement.git
   

3. Install dependencies:

   # For ComfyUI Portable (Windows)
   ../python_embeded/python.exe -m pip install -r ComfyUI-MemoryManagement/requirements.txt
   
   # For standard Python installation
   pip install -r ComfyUI-MemoryManagement/requirements.txt
   

4. Restart ComfyUI

Method 3: Manual Download

1. Download ZIP from GitHub repository
2. Extract to ComfyUI/custom_nodes/ComfyUI-MemoryManagement
3. Install dependencies (see Method 2, step 3)
4. Restart ComfyUI

⚙️ Configuration

Environment Variables (Optional)

# Set logging level
export COMFY_MEMORY_LOG_LEVEL=INFO

# Set memory thresholds
export COMFY_MEMORY_WARNING_THRESHOLD=75
export COMFY_MEMORY_CRITICAL_THRESHOLD=85

ComfyUI Launch Parameters

For systems with limited VRAM, use these launch parameters:

# Low VRAM mode
python main.py --lowvram

# CPU fallback
python main.py --cpu

# Memory optimization
python main.py --reserve-vram 2 --disable-smart-memory

🔍 Verification

Check Installation

1. Launch ComfyUI
2. Look for memory management nodes in the node menu under "Memory Management"
3. Check console for initialization messages:

   [ComfyUI-MemoryMgmt] INFO: ComfyUI integration enabled
   [ComfyUI-MemoryMgmt] INFO: Successfully loaded 7 memory management nodes
   

Test Basic Functionality

# Add a Memory Monitor node to your workflow
# Set refresh_trigger to 1
# Execute the workflow
# Check output for memory statistics

🏭 Production Settings

Recommended Node Configurations

Memory Monitor Node

• show_detailed: True (for debugging)
• show_gpu_info: True (if GPU available)
• memory_threshold_warning: 75.0

Smart Memory Manager

• management_mode: "Balanced"
• memory_warning_threshold: 75.0
• memory_critical_threshold: 85.0
• check_interval: 45.0
• enable_leak_detection: True

VRAM Optimizer

• optimization_level: "Moderate" (production)
• reset_peak_stats: True
• target_device: -1 (all devices)

Performance Optimization

# For high-throughput environments
Smart_Memory_Manager_Settings = {
    "check_interval": 30.0,  # More frequent checks
    "management_mode": "Aggressive",  # Maximum optimization
    "enable_vram_optimization": True,
    "auto_cleanup_aggressive": True
}

# For stability-focused environments  
Conservative_Settings = {
    "check_interval": 60.0,  # Less frequent checks
    "management_mode": "Conservative",
    "enable_vram_optimization": False,
    "auto_cleanup_aggressive": False
}

🔒 Security Considerations

File Permissions

# Ensure proper permissions
chmod 755 /path/to/ComfyUI/custom_nodes/ComfyUI-MemoryManagement
chmod 644 /path/to/ComfyUI/custom_nodes/ComfyUI-MemoryManagement/*.py

Network Security

• Memory management nodes operate locally only
• No network connections or external data transmission
• All operations are confined to the ComfyUI process

📊 Monitoring & Maintenance

Log Files

Check ComfyUI console output for memory management logs:

[ComfyUI-MemoryMgmt] INFO: Memory cleanup completed: 1543 objects collected
[ComfyUI-MemoryMgmt] WARNING: Critical memory usage: 87.3%
[ComfyUI-MemoryMgmt] INFO: VRAM optimization completed: 2.1GB freed

Performance Metrics

Monitor these key metrics:

• Memory Usage: Should stay below 85%
• Cleanup Frequency: Should not be too frequent (indicates issues)
• Leak Detection: Should report minimal leaks
• VRAM Utilization: Optimal usage without overflow

Maintenance Tasks

# Weekly: Check for updates
cd ComfyUI/custom_nodes/ComfyUI-MemoryManagement
git pull origin main

# Monthly: Verify dependencies
pip check

# As needed: Clear old snapshots (handled automatically)

🚨 Troubleshooting

Common Issues

Nodes Not Appearing

# Check dependencies
pip install psutil torch

# Check console for errors
# Look for import errors or missing dependencies

Memory Management Not Working

# Verify CUDA availability (for VRAM optimization)
python -c "import torch; print(torch.cuda.is_available())"

# Check system permissions
# Ensure ComfyUI has sufficient privileges

Performance Issues

# Reduce monitoring frequency
# Use Conservative mode instead of Aggressive
# Disable leak detection if not needed

Error Codes

• CUDA not available: Install CUDA toolkit
• Missing dependencies: Run pip install -r requirements.txt
• Permission denied: Check file permissions
• Memory monitoring error: Check system resources

🔧 Advanced Configuration

Custom Thresholds

# In your workflow, configure Smart Memory Manager:
warning_threshold = 70.0  # Start cleanup at 70%
critical_threshold = 80.0  # Aggressive cleanup at 80%
check_interval = 20.0     # Check every 20 seconds

Integration with Other Tools

# For use with other monitoring tools
# Memory data can be accessed via node outputs
memory_data = memory_monitor_output
# Process with external monitoring systems

📞 Support

Getting Help

1. Check logs in ComfyUI console
2. Review this documentation
3. Search existing issues on GitHub
4. Create detailed bug report with:
   • System specifications
   • ComfyUI version
   • Error messages
   • Reproduction steps

Community Resources

• ComfyUI Discord: Memory Management channel
• GitHub Issues: Bug reports and feature requests
• ComfyUI Forum: Community discussions

---

✅ Deployment Checklist

• ComfyUI installed and working
• Dependencies installed (psutil, torch)
• Memory management nodes visible in ComfyUI
• Console shows successful initialization
• Basic memory monitoring working
• VRAM optimization functional (if GPU available)
• Smart memory manager configured
• Performance thresholds set appropriately
• Monitoring and logging configured
• Backup and update procedures established

🎉 Your ComfyUI Memory Management system is now production-ready!