CloudAI Fusion

Cloud-Native AI Unified Management Platform

<a href="https://github.com/cloudai-fusion/cloudai-fusion/actions"><img src="https://github.com/cloudai-fusion/cloudai-fusion/workflows/CloudAI%20Fusion%20CI%2FCD/badge.svg" alt="CI"></a>
<a href="https://codecov.io/gh/cloudai-fusion/cloudai-fusion"><img src="https://codecov.io/gh/cloudai-fusion/cloudai-fusion/branch/main/graph/badge.svg" alt="Coverage"></a>
<a href="https://goreportcard.com/report/github.com/cloudai-fusion/cloudai-fusion"><img src="https://goreportcard.com/badge/github.com/cloudai-fusion/cloudai-fusion" alt="Go Report"></a>
<a href="https://github.com/QQ3221197721/cloudai-fusion/raw/main/LICENSE"><img src="https://img.shields.io/badge/License-Apache%202.0-blue.svg" alt="License"></a>
<a href="https://github.com/cloudai-fusion/cloudai-fusion/releases"><img src="https://img.shields.io/github/v/release/cloudai-fusion/cloudai-fusion" alt="Release"></a>

CloudAI Fusion is an open-source platform that unifies cloud-native infrastructure management with AI-powered resource scheduling. It solves three core enterprise pain points:

• Cloud-native deployment complexity — 68% of enterprises struggle with cluster management
• Multi-cloud security fragmentation — 86% of organizations use multi-cloud but face siloed security
• AI resource waste — GPU utilization under 30% in traditional deployment models

Key Features

Architecture

┌─────────────────────────────────────────────────────────────────────┐
│                         CloudAI Fusion                              │
├─────────────┬──────────────┬─────────────┬─────────────────────────┤
│  API Server │  Scheduler   │    Agent    │      AI Engine          │
│   (Go/Gin)  │ (GPU-aware)  │ (DaemonSet) │   (Python/FastAPI)     │
├─────────────┴──────────────┴─────────────┴─────────────────────────┤
│  Auth │ Cloud │ Cluster │ Security │ Monitor │ Mesh │ Wasm │ Edge  │
├───────┴───────┴─────────┴──────────┴─────────┴──────┴──────┴───────┤
│            PostgreSQL  │  Redis  │  Kafka  │  NATS  │  Prometheus     │
└─────────────────────────────────────────────────────────────────────┘

Quick Start

Prerequisites

• Docker Desktop (with Docker Compose)

Run (one command)

Windows:

cd cloudai-fusion
start.bat

Linux / macOS:

cd cloudai-fusion
chmod +x start.sh && ./start.sh

This starts all services (API Server, Scheduler, Agent, AI Engine, PostgreSQL, Redis, Kafka, NATS, Prometheus, Grafana, Jaeger).

Verify

# Health check
curl http://localhost:8080/healthz

# Login (get JWT token)
curl -X POST http://localhost:8080/api/v1/auth/login \
  -H "Content-Type: application/json" \
  -d '{"username":"admin","password":"admin123"}'

# List clusters (use token from login response)
curl http://localhost:8080/api/v1/clusters \
  -H "Authorization: Bearer <token>"

Service Endpoints

Tech Stack

Project Structure

cloudai-fusion/
├── cmd/                          # Service entry points
│   ├── apiserver/                # API Server
│   ├── scheduler/                # GPU Scheduler
│   ├── agent/                    # Node Agent
│   └── healthcheck/              # Lightweight healthcheck binary (for distroless)
├── pkg/                          # Go packages (48 packages, 47/47 tests pass)
│   ├── api/                      # HTTP routes, handlers, debug endpoints (6-layer auth)
│   ├── auth/                     # JWT + RBAC + OIDC federation (4 roles, 20+ perms)
│   ├── cache/                    # Redis cache + distributed lock + PubSub
│   ├── cloud/                    # Multi-cloud providers (6 clouds)
│   ├── cluster/                  # Kubernetes cluster management
│   ├── config/                   # Unified configuration + JWT secret validation
│   ├── edge/                     # Edge-cloud architecture + 50B model quantization
│   ├── feature/                  # Runtime feature toggles (minimal/standard/full)
│   ├── mesh/                     # eBPF service mesh
│   ├── monitor/                  # Prometheus metrics & alerting
│   ├── scheduler/                # GPU scheduling + queue snapshot persistence
│   ├── security/                 # Policy, scanning, compliance, threats
│   ├── store/                    # Database layer (GORM)
│   └── wasm/                     # WebAssembly container runtime
├── ai/                           # Python AI components
│   ├── agents/                   # Multi-agent FastAPI server + LLM client
│   ├── anomaly/                  # Anomaly detection engine
│   └── scheduler/                # RL scheduling optimizer
├── deploy/helm/                  # Helm chart (7 templates)
├── docker/                       # Dockerfiles (5 services, distroless + multi-stage)
├── monitoring/                   # Prometheus + Grafana configs
├── scripts/                      # env-generate, diagnose, setup-gpu, deploy-cloud
├── api/                          # OpenAPI 3.1 specification
├── .devcontainer/                # Dev Container (Go + Python + full toolchain)
├── docker-compose.yml            # Full-stack local deployment (profiles support)
├── docker-compose.gpu.yml        # GPU overlay (NVIDIA runtime)
├── Makefile                      # Build, test, deploy, diagnose commands
└── start.bat / start.sh          # One-click start (--gpu, --minimal flags)

Development

# First-time setup (generate .env with secure secrets)
make setup

# Build all Go binaries
make build

# Run tests
make test

# Run tests with coverage report
make coverage-report

# Run linter
make lint

# Build Docker images (optimized: distroless for Go, multi-stage for AI)
make docker-build

# Deploy to Kubernetes
make helm-install

# Start with GPU support
./start.sh --gpu          # Linux / macOS
start.bat --gpu           # Windows (WSL2 + NVIDIA)

# Start minimal mode (core services only)
./start.sh --minimal

# Health diagnostics
make diagnose

# Feature flag management
make features-list

API Overview

Full specification: api/openapi.yaml

Roadmap

Contributing

See CONTRIBUTING.md for guidelines.

License

Apache License 2.0