Add multi-GPU support for Docker deployments

- Created separate docker-compose override files for different GPU types:
  - docker-compose.nvidia.yml for NVIDIA GPUs
  - docker-compose.amd.yml for AMD GPUs with ROCm
  - docker-compose.apple.yml for Apple Silicon
- Updated README with GPU-specific Docker configurations
- Updated deployment instructions to use appropriate override files
- Added detailed configurations for each GPU type including:
  - Device mappings and drivers
  - Environment variables
  - Platform specifications
  - Memory and resource limits

This allows users to easily deploy Talk2Me with their specific GPU hardware.

🤖 Generated with [Claude Code](https://claude.ai/code)

Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
2025-06-03 09:16:41 -06:00
parent e5333d8410
commit bcbac5c8b3
4 changed files with 138 additions and 3 deletions

19
docker-compose.amd.yml Normal file
View File

@@ -0,0 +1,19 @@
version: '3.8'
# Docker Compose override for AMD GPU support (ROCm)
# Usage: docker-compose -f docker-compose.yml -f docker-compose.amd.yml up
services:
talk2me:
environment:
- HSA_OVERRIDE_GFX_VERSION=10.3.0 # Adjust based on your GPU model
- ROCR_VISIBLE_DEVICES=0 # Use first GPU
volumes:
- /dev/kfd:/dev/kfd # ROCm KFD interface
- /dev/dri:/dev/dri # Direct Rendering Interface
devices:
- /dev/kfd
- /dev/dri
group_add:
- video # Required for GPU access
- render # Required for GPU access