Skip to main content
Version: 1.1

Architecture Overview

System Architecture​

Secan follows a modern client-server architecture designed for efficient Elasticsearch cluster management.

Component Breakdown​

Frontend (React)​

  • Dashboard: Visual overview of cluster health and metrics
  • Cluster Management: View and manage cluster settings
  • Index Management: Create, delete, and configure indices
  • Shard Management: Monitor and manage shard allocation
  • REST Console: Interactive API exploration tool

Backend (Rust)​

  • HTTP Server: High-performance Actix-web server
  • Authentication: Support for multiple auth methods (Basic, OAuth, API Keys)
  • Request Routing: Intelligent routing to appropriate handlers
  • Caching: Intelligent caching for frequently accessed data
  • API Proxying: Secure communication with Elasticsearch clusters

Elasticsearch Cluster​

  • Nodes: Multiple nodes for redundancy and scalability
  • Shards: Distributed data storage and processing
  • Replicas: Data redundancy and fault tolerance
  • Indices: Organized data structures for efficient querying

Data Flow​

  1. User Interaction: User interacts with React frontend
  2. HTTP Request: Frontend sends HTTP request to backend
  3. Authentication: Backend validates user credentials/tokens
  4. Request Processing: Backend routes request to appropriate handler
  5. Elasticsearch Query: Backend queries Elasticsearch cluster
  6. Response Processing: Backend processes and caches response
  7. UI Update: Frontend receives response and updates UI

Security Architecture​

  • Authentication Layer: Protects backend endpoints
  • Encryption: HTTPS for data in transit
  • Token-based Auth: Support for JWT and API keys
  • Request Validation: Input validation on all endpoints
  • Audit Logging: Track all administrative actions

Scalability Features​

  • Stateless Backend: Easy horizontal scaling
  • Connection Pooling: Efficient resource utilization
  • Request Caching: Reduced load on Elasticsearch
  • Multi-cluster Support: Manage multiple Elasticsearch clusters
  • Async Processing: Non-blocking operations for better throughput