AI/Run CodeMie Deployment Architecture

The diagram below depicts the AI/Run CodeMie infrastructure deployment in one region (AZ) of the AWS public cloud environment.

Container Resources Requirements

Component	Pods	RAM	vCPU	Storage
CodeMie API	2	8Gi	4.0	–
CodeMie UI	1	128Mi	0.1	–
Elasticsearch	2	16Gi	4.0	100-200 Gb per pod
Kibana	1	1Gi	1.0	–
Mermaid-server	1	512Mi	1.0	–
PostgreSQL	Managed service in cloud			30-50 Gb
Keycloak + DB	1 + 1	4Gi	2.0	1 Gb
Oauth2-proxy	1	128Mi	0.1	–
NATS + Auth Callout	1 + 1	512Mi	1.0	–
MCP Connect	1	1Gi	0.5	–
Fluent Bit	daemonset	128Mi	0.1	–
LLM Proxy*	1	1Gi	1.0	–

*Depends on the exact LLM proxy type

Infrastructure Components

The AI/Run CodeMie deployment on AWS includes the following main infrastructure components:

Compute

• Amazon EKS Cluster: Managed Kubernetes service for running containerized applications
• Auto Scaling Groups (ASG): Automatically adjusts compute capacity for the EKS cluster

Networking

• VPC: Isolated network environment
• Application Load Balancer (ALB): Distributes incoming HTTPS traffic to application services
• Network Load Balancer (NLB): Handles TCP traffic for NATS messaging system
• NAT Gateway: Provides outbound internet connectivity for private subnets
• Route 53: DNS service with automatically provisioned records

Storage

• Amazon RDS PostgreSQL: Managed relational database service
• Amazon S3: Object storage for application data and artifacts
• EBS Volumes: Persistent block storage for Kubernetes pods

Security

• AWS KMS: Key Management Service for encrypting and decrypting sensitive data
• IAM Roles: Granular permissions for services and components
• AWS Certificate Manager: Automated SSL/TLS certificate management for ALB and NLB
• Security Groups: Network-level access control

Optional Features

• Internal ALB: For private network communication when enabled
• Private DNS Hosted Zone: For internal service discovery