Skip to content

ARCHITECTURE.md

Latest commit

 

History

History
118 lines (77 loc) · 3.17 KB

ARCHITECTURE.md

File metadata and controls

118 lines (77 loc) · 3.17 KB
Raw

🏗️ hexaFn Architecture

⚙️ Core Principle: The 6F Lifecycle Flow

hexaFn is architected around a six-phase functional data pipeline called the 6F Lifecycle Flow:

Feed → Filter → Format → Function → Forward → Feedback

Each phase in the flow is programmable, observable, and independently testable. Modules are designed to be loosely coupled and highly composable.

🔶 Module Overview

1. HexaBridge

  • Integration module: external API/webhook/SDK interfaces
  • Connects hexaFn to the outside world
  • Enables bidirectional communication

2. HexaTrigger

  • Conditional logic and rule engine
  • Event chaining with AND/OR logic
  • Time-based and value-based triggering supported

3. HexaRun

  • Runtime for dynamic function execution
  • Supports: WASM, Lua, JS, internal DSL
  • Can access KV, cast, and external inputs in a sandboxed context

4. HexaStore

  • Typed, schema-validating key-value store
  • Supports triggers, TTL, event emit on write
  • Backend-agnostic (supports in-memory, disk-based, or Redis-style engines)

5. HexaCast

  • High-performance Pub/Sub engine
  • Topic-based + pattern matching subscriptions
  • Delivers messages to functions, external services, or UI listeners

6. HexaWatch

  • Unified observer layer (logging, tracing, metrics)
  • Supports pluggable sinks (console, Elastic, Prometheus)
  • Real-time audit trail of system activity

🧠 Data Flow (Simplified Example)

6F Data Flow Diagram

User Event → Feed (Ingest) 
  → Filter (Check validity or user type) 
    → Format (Shape into canonical JSON) 
      → Function (Calculate next action) 
        → Forward (Notify via pubsub + store update) 
          → Feedback (Log + trigger follow-up)

All stages can independently be observed, tested, logged, and extended.

🧩 Example Use Cases

Real-time Chat Moderation

  • Incoming chat messages are fed into the system.
  • Messages are filtered for profanity and spam.
  • Formatted to standard JSON structure.
  • Functions apply sentiment analysis and auto-moderation rules.
  • Forwarded alerts sent to moderators via pubsub.
  • Feedback logs moderation actions for audit.

IoT Device Data Processing

  • Devices send sensor data (temperature, humidity).
  • Feed module ingests device events.
  • Filter discards invalid or out-of-range readings.
  • Format normalizes units and data formats.
  • Function computes trends and anomaly detection.
  • Forward sends processed data to dashboard APIs.
  • Feedback stores audit trail and triggers alerts if needed.

🏗️ Deployment Flexibility

  • Fully self-hostable via Docker or bare-metal
  • CLI and REST control surfaces (coming soon)
  • Future support for distributed mode & scaling workers

📦 File Structure (Planned)

➡️ PROJECT_STRUCTURE – Directory structure and module breakdown

📬 Questions?

Contact the team at info@hexafn.com or open an issue.

Built by hexaTune LLC · GitHub: hTuneSys/hexaFn