utility-backend

Enterprise utility telemetry ingestion, tariff evaluation, and blockchain settlement backend.

Architecture

src/
├── gateway/     - mTLS, gRPC, MQTT hooks for hardware utility meters
├── tariffs/     - Dynamic temporal/volumetric pricing logic
├── time_series/ - TimescaleDB ingestion & analytics pipelines
├── soroban/     - Soroban RPC batch settlement transactions
└── api/         - Protected dashboard & credential endpoints

Quick Start

docker compose up -d

Development

cargo test --all-features
cargo clippy --all-targets -- -D warnings

CI/CD

GitHub Actions runs lint, type-check, and Dockerized database tests on every commit.

Capacity Planning

The backend exposes a system-wide capacity planning forecast at GET /api/v1/capacity/forecast. The planner groups historical usage samples by service and resource, computes a linear utilization trend, and returns the current utilization, projected utilization over the default 30-day horizon, estimated days to warning/critical thresholds, and an operational recommendation.

Prometheus metrics are emitted for dashboarding and alerting:

• utility_capacity_current_utilization_ratio{service,resource}
• utility_capacity_projected_utilization_ratio{service,resource}
• utility_capacity_days_to_critical{service,resource}

Suggested alerts:

• Page when projected utilization reaches the critical threshold inside the planning horizon.
• Warn when current or projected utilization exceeds the warning threshold.
• Review runbooks before blue-green promotion when any resource returns scale_immediately or scale_within_horizon.