Create Grafana dashboard template for metrics visualization

[0xReLogic]

Description

Create a Grafana dashboard template (JSON) to visualize Helios metrics exposed at /metrics endpoint.

Requirements

• Grafana dashboard JSON file
• Panels for key metrics:
  • Request rate (RPS)
  • Response latency (p50, p95, p99)
  • Backend health status
  • Active connections count
  • Circuit breaker state
  • Rate limit metrics
• Variables for backend selection
• Time range controls

Deliverables

1. grafana/helios-dashboard.json - Dashboard template
2. grafana/README.md - Import instructions
3. Screenshots in documentation
4. Update main README with monitoring section

Dashboard Panels

• Overview: Total requests, error rate, avg latency
• Backends: Per-backend RPS, latency, health status
• Circuit Breaker: State timeline, trip events
• Rate Limiting: Tokens consumed, rejected requests
• System: Memory usage, goroutines, CPU

Acceptance Criteria

• Complete Grafana dashboard JSON
• Import instructions with screenshots
• Documentation in README
• Tested with Prometheus data source

References

• https://grafana.com/docs/grafana/latest/dashboards/json-model/
• Current metrics endpoint: :9090/metrics

[trivedi-khushi]

Hey,
I work as a Technical Documentation writer for my company and have worked for the Guides, User-Docs, API References & SDK guides. I wish to contribute to this issue, and would like to share my professional expertise on the same.

Kindly assign this issue to me @0xReLogic

[0xReLogic]

Hi @trivedi-khushi,

Thank you for your interest in contributing to Helios!

I'm assigning this issue to you. Here are some important details:

Current Metrics API:
The /metrics endpoint (port 9090) returns JSON with the following structure:

{
  "total_requests": 2410,
  "successful_requests": 2410,
  "failed_requests": 0,
  "total_response_time_ms": 13652,
  "average_response_time_ms": 5.66,
  "backend_metrics": {
    "server1": {
      "name": "server1",
      "total_requests": 803,
      "successful_requests": 803,
      "failed_requests": 0,
      "active_connections": 0,
      "average_response_time_ms": 5.28
    }
  },
  "rate_limited_requests": 0,
  "circuit_breaker_metrics": {},
  "start_time": "2025-08-05T11:33:02+08:00",
  "uptime": "2m13s"
}

[0xReLogic]

Dashboard Suggestions:

• Use JSON API data source (not Prometheus) since our metrics are JSON-based
• Create panels for RPS, latency percentiles, error rates
• Add backend comparison panels
• Include rate limiter and circuit breaker status
• Note: Since we use JSON metrics (not Prometheus format), you may need to document using Grafana's JSON API data source plugin or consider adding a Prometheus exporter in the future.

Feel free to ask if you need sample data or have questions about the metrics format!

Looking forward to your contribution.