Skip to main content

​
Production Deployment Guide

Deploy Shannon to your infrastructure with confidence. This section covers deployment patterns, cloud platform integrations, and operational best practices.
Shannon is currently in active development. Production deployment guides are being finalized. For production use, we recommend:
  • Thorough testing in staging environments
  • Monitoring all services closely
  • Joining our Discord for deployment support

​
Deployment Options

Docker Compose

Production-ready Docker Compose configuration
Status: 🚧 Phase 3

Kubernetes

Kubernetes manifests and Helm charts
Status: 🚧 Phase 3

AWS

Deploy to Amazon Web Services (ECS, RDS, ElastiCache)
Status: 🚧 Phase 3

Azure

Deploy to Microsoft Azure (AKS, PostgreSQL, Redis)
Status: 🚧 Phase 3

​
Operations

Monitoring

Prometheus metrics, Grafana dashboards, and alerting
Status: 🚧 Phase 3

Performance Tuning

Optimize throughput, latency, and resource usage
Status: 🚧 Phase 3

Security

Production security hardening and best practices
Status: 🚧 Phase 3

​
Quick Start: Local Development

For development and testing, use Docker Compose: 
# Clone repository
git clone https://github.com/Kocoro-lab/Shannon.git
cd Shannon

# Setup configuration
make setup
echo "OPENAI_API_KEY=sk-your-key-here" >> .env

# Download Python WASI interpreter
./scripts/setup_python_wasi.sh

# Start all services
make dev
See Installation Guide for detailed setup instructions.

​
Architecture Overview

Shannon consists of multiple services that need to be deployed:

​
Core Services

ServicePurposeScaling
GatewayREST API, authenticationHorizontal (stateless)
OrchestratorTask coordination, gRPCHorizontal (stateful via Temporal)
Agent CoreAgent execution, Rust runtimeHorizontal
LLM ServiceLLM provider gatewayHorizontal
DashboardReal-time monitoring UIHorizontal (stateless)

​
Data Stores

StorePurposeScaling
PostgreSQLTask metadata, events, sessionsVertical + read replicas
RedisCaching, pub/sub, sessionsCluster mode
QdrantVector embeddings, semantic memoryHorizontal
TemporalWorkflow state, durable executionCluster mode

​
Production Checklist

Before deploying to production:

​
Security

  • Enable authentication (GATEWAY_SKIP_AUTH=0)
  • Configure TLS/SSL for all services
  • Rotate API keys regularly
  • Set up OPA policies for access control
  • Enable audit logging
  • Configure network policies/firewalls

​
Reliability

  • Set up health checks and readiness probes
  • Configure auto-scaling policies
  • Implement circuit breakers
  • Set resource limits (CPU, memory)
  • Configure backup and disaster recovery
  • Test failover scenarios

​
Observability

  • Deploy Prometheus and Grafana
  • Configure alerting rules
  • Set up log aggregation (ELK/Loki)
  • Enable distributed tracing (OpenTelemetry)
  • Create runbooks for common issues

​
Performance

  • Tune Temporal worker concurrency
  • Optimize database connections
  • Configure Redis caching
  • Set appropriate resource limits
  • Load test before production launch

​
Resource Requirements

​
Minimum (Development)

  • CPU: 4 cores
  • RAM: 8GB
  • Storage: 20GB SSD
  • CPU: 16 cores total (distributed across services)
  • RAM: 32GB total
  • Storage: 100GB SSD
  • Network: 1Gbps
  • CPU: 64+ cores
  • RAM: 128GB+
  • Storage: 500GB+ SSD
  • Network: 10Gbps
  • Load Balancer: Required
  • Multi-AZ: Recommended

​
What’s Next?

Quick Start

Install Shannon locally first

Configuration

Understand environment variables

Architecture

Learn system architecture

Monitoring

Set up monitoring

​
Get Help

  • Discord: Join our community for deployment help
  • GitHub: File deployment issues or questions
  • Docs: Check Troubleshooting for common problems