Skip to main content

Gateway Architecture

The Gateway serves as the primary entry point and orchestration layer for Lagrange's ZK Prover Network. It provides a unified interface for clients while managing the complex distributed proving infrastructure behind the scenes.

Core Responsibilities​

Client Interface Management​

The Gateway provides a standardized API layer that:

  • Authenticates and authorizes incoming proof requests
  • Normalizes inputs across different proof systems and client implementations
  • Validates request parameters to ensure correctness before processing
  • Provides real-time status updates to clients about proof generation progress

Task Orchestration​

Once a request is received, the Gateway:

  • Queues tasks persistently to ensure no requests are lost
  • Breaks down complex proofs into manageable job slices
  • Coordinates job distribution across available provers
  • Manages dependencies between related proof generation tasks

Proof Aggregation​

The Gateway handles the final assembly of proofs by:

  • Collecting partial proofs from individual provers
  • Verifying proof correctness before aggregation
  • Combining proofs into the final output format
  • Streaming results back to the requesting client

Request Processing Pipeline​

  1. Input Validation: Verify request format and parameters
  2. Authentication: Validate client credentials and permissions
  3. Rate Limiting: Enforce usage quotas and prevent abuse
  4. Request Queuing: Persist tasks for reliable processing

Dispatcher Integration​

The Gateway contains an embedded Dispatcher component that:

  • Analyzes job requirements (computational complexity, hardware needs)
  • Classifies jobs into small, medium, or large categories
  • Interfaces with DARA for optimal prover selection
  • Monitors job progress and handles failures

Storage Integration​

  • Task Persistence: Reliable queuing with durability guarantees
  • Result Caching: Efficient storage and retrieval of proof outputs
  • Metadata Tracking: Comprehensive logging of all operations
  • Backup Systems: Redundant storage for critical data

Fault Tolerance & Reliability​

Automatic Retry Logic​

The Gateway implements sophisticated retry mechanisms:

  • Exponential Backoff: Intelligent retry timing to avoid overwhelming provers
  • Prover Reassignment: Automatic failover to alternative provers
  • Partial Result Recovery: Resume from intermediate states when possible
  • Timeout Management: Configurable timeouts with graceful degradation

High Availability Design​

  • Multi-Region Deployment: Geographic distribution for global access
  • Load Balancing: Distribute requests across multiple Gateway instances
  • Health Monitoring: Continuous monitoring of system components
  • Graceful Degradation: Maintain service during partial outages

Performance Optimization​

Efficient Resource Utilization​

  • Connection Pooling: Reuse connections to reduce overhead
  • Batch Processing: Group similar requests for efficiency
  • Caching Strategies: Cache frequently accessed data and results
  • Asynchronous Processing: Non-blocking operations for better throughput

Scalability Features​

  • Horizontal Scaling: Add more Gateway instances as demand grows
  • Dynamic Load Balancing: Adjust traffic distribution based on capacity
  • Resource Monitoring: Track and optimize resource usage
  • Predictive Scaling: Anticipate demand and pre-scale resources

Security Model​

Data Protection​

  • End-to-End Encryption: Secure data transmission and storage
  • Input Sanitization: Prevent injection attacks and malformed data
  • Secure Communication: TLS/SSL for all network communications
  • Data Isolation: Separate client data and processing contexts

Network Security​

  • DDoS Protection: Mitigation of distributed denial-of-service attacks
  • Rate Limiting: Prevent abuse and ensure fair resource allocation
  • Firewall Integration: Network-level security controls
  • Intrusion Detection: Monitor for suspicious activities

Integration Patterns​

Application-Specific Subnetworks​

  • Dedicated Resources: Reserved proving capacity for high-volume applications
  • Custom SLAs: Tailored service level agreements
  • Specialized Hardware: Optimized prover configurations
  • Priority Queuing: Guaranteed processing order for critical applications

Monitoring & Observability​

Real-Time Metrics​

  • Request Throughput: Track requests per second and processing times
  • Success Rates: Monitor proof generation success and failure rates
  • Resource Utilization: CPU, memory, and network usage statistics

Alerting System​

  • Threshold-Based Alerts: Automated notifications for anomalies
  • Escalation Procedures: Structured response to critical issues
  • Integration Support: Connect with monitoring tools
  • Custom Dashboards: Tailored views for different stakeholders

Performance Analytics​

  • Historical Trends: Long-term performance and usage patterns
  • Cost Optimization: Identify opportunities for efficiency improvements
  • User Behavior Analysis: Understand client usage patterns