Cloud-Native K8s Elastic Scaling Architecture

System Design · flowchart diagram · Apache-2.0

A boundary-first layered architecture for cloud-native Kubernetes, separating ingress, control plane, runtime orchestration, and model access for enhanced

Source: https://github.com/machinesociety/ClawLoops/blob/7d2c1214f45e98d01c1e3dc8b8c519a1c2e53488/README/README_ko-KR.md
Curated by machinesociety
Kubernetes Cloud-Native Traefik PostgreSQL LLM System Design Scalability

Mermaid source

flowchart TB
    A([Browser])

    subgraph L1[인그레스 계층]
        direction TB
        B[Traefik]
    end

    subgraph L2[플랫폼 계층]
        direction TB
        D[ClawLoops Control Plane]
    end

    subgraph L3[오케스트레이션 계층]
        direction TB
        E[Runtime Orchestrator]
        F[Runtime Manager]
    end

    subgraph L4[런타임 계층]
        direction TB
        G[Per-user OpenClaw Runtime]
    end

    subgraph L5[모델 계층]
        direction TB
        H[Model Service / Credential Proxy]
        I[LiteLLM]
        J[vLLM / Ollama / Upstream Providers]
    end

    K[(PostgreSQL)]

    A -->|Ingress| B -->|Session check| D
    D -->|Desired state| E -->|Lifecycle orchestration| F -->|Start / Stop / Recover| G
    D -->|Model access| H --> I --> J
    D -->|Platform data| K

    classDef edge fill:#eff6ff,stroke:#2563eb,stroke-width:2px,color:#0f172a;
    classDef plane fill:#dcfce7,stroke:#16a34a,stroke-width:2px,color:#0f172a;
    classDef runtime fill:#f3e8ff,stroke:#7c3aed,stroke-width:2px,color:#0f172a;
    classDef model fill:#ffe4e6,stroke:#db2777,stroke-width:2px,color:#0f172a;
    classDef data fill:#f8fafc,stroke:#64748b,stroke-width:2px,color:#0f172a;

    class A,B edge;
    class D plane;
    class E,F,G runtime;
    class H,I,J model;
    class K data;

What this diagram shows

This diagram illustrates a multi-layered cloud-native Kubernetes architecture for elastic scaling. It details five distinct layers: Ingress (Traefik), Platform (ClawLoops Control Plane), Orchestration (Runtime Orchestrator, Runtime Manager), Runtime (Per-user OpenClaw Runtime), and Model (Model Service, LiteLLM, vLLM/Ollama/Upstream Providers). It also shows data flow between these layers and a PostgreSQL database for platform data, emphasizing session checks, lifecycle orchestration, and model access paths.

When to use it

Use this architecture when designing highly scalable and secure cloud-native applications on Kubernetes, particularly for multi-tenant environments or platforms that require dynamic provisioning of user-specific runtimes and integration with various AI/ML model services. It's suitable for scenarios demanding clear separation of concerns, robust access control, and efficient resource orchestration.

How to adapt it for your project

This architecture can be adapted by swapping specific components like the ingress controller (e.g., Nginx Ingress), customizing the control plane logic for different business needs, integrating alternative runtime environments (e.g., serverless functions), or connecting to different model providers. The layered design allows for independent scaling and evolution of each component, enabling flexibility in technology choices and operational strategies.

Key concepts