This directory contains Ansible playbooks for deploying OptiReduce and its dependencies. For detailed information, visit our official documentation.
- 📥 Download
- 🔑 Prerequisites
- 📂 Directory Structure
- ⚙️ Configuration
- 🚀 Deployment Options
- 🧩 Available Components
- 🌍 Environment Variables
⚠️ Troubleshooting- 📚 Additional Resources
- 🆘 Support
- 📜 License
Clone the Ansible repository:
git clone https://github.com/OptiReduce/ansible.git
cd ansibleThe playbooks automate the deployment of all OptiReduce components.
Ubuntu/Debian:
sudo apt update
sudo apt install software-properties-common
sudo apt-add-repository --yes --update ppa:ansible/ansible
sudo apt install ansibleRHEL/CentOS:
sudo yum install epel-release
sudo yum install ansibleVerify installation:
ansible --version#!/bin/bash
sudo apt update && sudo apt install -y openssh-server
sudo systemctl enable ssh
sudo systemctl start ssh
sudo systemctl status ssh --no-pagerUsage:
chmod +x install_ssh.sh
./install_ssh.sh#!/bin/bash
if [ "$#" -ne 2 ]; then
echo "Usage: $0 <target_username> <target_host>"
exit 1
fi
TARGET_USER="$1"
TARGET_HOST="$2"
SSH_KEY="$HOME/.ssh/id_rsa"
[ ! -f "$SSH_KEY" ] && ssh-keygen -t rsa -b 4096 -N "" -f "$SSH_KEY"
ssh-copy-id "$TARGET_USER@$TARGET_HOST"
ssh -o BatchMode=yes "$TARGET_USER@$TARGET_HOST" "echo 'SSH connection successful on $(hostname)!'"Steps:
- Save as
ssh_setup.shand make it executable:chmod +x ssh_setup.sh
- Run with target credentials:
./ssh_setup.sh user 192.168.1.10
optireduce/
├── ansible.cfg # Ansible configuration
├── inventory/
│ └── hosts # Target machine definitions
├── group_vars/
│ └── all.yml # Global variables
├── optireduce_deploy.yml # Main playbook
├── Makefile # Deployment shortcuts
└── roles/ # Component roles
├── cuda/ # CUDA 11.7 setup
├── mellanox/ # Mellanox drivers
├── anaconda/ # Python environment
├── optireduce/ # Core OptiReduce
└── benchmark/ # Benchmark tools
[gpu_nodes]
node1 ansible_host=192.168.1.101 ansible_user=test
node2 ansible_host=192.168.1.102 ansible_user=testcuda_version: "11.7.0-1"
nvidia_version: "515"
cudnn_version: "8.5.0.96-1+cuda11.7"
python_version: "3.9.19"
dpdk_version: "v20.11"| Command | Description |
|---|---|
make optireduce-full |
Full installation |
make cuda-only |
Install CUDA only |
make benchmark-only |
Install benchmarks |
make check |
Validate configuration |
Custom Installation:
make deploy INSTALL_CUDA=true INSTALL_BENCHMARK=true- CUDA 11.7 with cuDNN 8.5
- Mellanox OFED Drivers
- Anaconda (Python 3.9.19)
- DPDK v20.11
- OptiReduce Core
- Benchmarking Tools
# Toggle components during deployment
INSTALL_CUDA=true/false
INSTALL_MELLANOX=true/false
INSTALL_ANACONDA=true/false
INSTALL_OPTIREDUCE=true/false
INSTALL_BENCHMARK=true/false| Issue | Solution |
|---|---|
| SSH Connection | Verify keys and network connectivity |
| CUDA Failures | Check NVIDIA repo access and space |
| OFED Errors | Confirm kernel compatibility |
- Check the Troubleshooting section above.
- Review Ansible logs at
/var/log/ansible.log. - Open an issue in the GitHub repository.
This deployment code is part of the OptiReduce project. Refer to the project page for licensing details.
### Key Enhancements:
1. **Badges**: Added Ansible, CUDA, and License badges for quick visual cues.
2. **Emojis**: Used emojis in headers (e.g., 📥, 🔑, ⚙️) to improve scannability.
3. **Syntax Highlighting**: All code blocks tagged with `bash`, `yaml`, `ini`, etc., for proper GitHub rendering.
4. **Tables**: Structured deployment options and troubleshooting as tables.
5. **Directory Structure**:```markdown Added comments to explain each file/folder.
6. **Consistent Formatting**: Separated sections with `---` lines and used bold text for emphasis.
7. **Links**: Hyperlinked documentation resources with emojis for clarity.