**Autonomous Resource Monitor **

An autonomous, self-aware infrastructure management platform designed to monitor system metrics, detect anomalies using statistical analysis, and provide real-time operational insights via a professional monitoring stack.

This project demonstrates a complete microservice ecosystem built using modern DevOps and ML practices, from data ingestion and analysis to visualization and automated lifecycle management.

---

Core Features

Microservice Architecture

A decoupled system consisting of:

• Backend Service (Spring Boot): Handles API logic and data storage.
• ML Worker Service (Python): Performs anomaly detection.
• Monitoring Stack (Prometheus + Grafana): Provides observability and analytics.

Real-Time Metric Ingestion

Spring Boot backend exposes REST APIs for collecting any time-series metric data, stored in a PostgreSQL database.

�Autonomous Anomaly Detection

A continuously running Python ML worker fetches recent metrics and applies the Modified Z-Score algorithm — a robust statistical method that identifies outliers without assuming a Gaussian distribution.

Observability Stack

Integrates Prometheus for metrics scraping and Grafana for interactive dashboards that visualize:

• System health
• Metric trends
• Detected anomalies

Containerized & Automated Deployment

All components — Backend, ML Worker, Database, Prometheus, Grafana — are containerized using Podman. A single automation script (run_project.sh) manages:

• Environment cleanup
• Image rebuild
• Sequential startup

This ensures reproducibility, consistency, and zero manual setup.

---

Technology Stack

Layer	Technology/Tool	Purpose
Backend	Java 17+, Spring Boot, Spring Data JPA, Spring Actuator	API, persistence, and metric exposure
ML/AI	Python, Pandas	Data manipulation and statistical anomaly detection
Database	PostgreSQL	Storage for time-series metrics
DevOps	Podman, Dockerfile, Bash, SELinux	Containerization and lifecycle automation
Monitoring	Prometheus, Grafana	Metrics collection and dashboard visualization
OS	Rocky Linux	Base runtime environment

---

Getting Started

Prerequisites

• Linux VM (tested on Rocky Linux 9)
• git, podman (or docker)
• Internet access for container image pulls

Installation

git clone https://github.com/prodXCE/autonomous-resource-optimizer
cd autonomous-resource-optimizer
chmod +x runAro.sh

Start the Stack

./runAro.sh

This script automatically:

• Stops existing containers
• Rebuilds images
• Launches services in order

---

Verification

Check Running Containers

sudo podman ps

You should see:

• aro-backend-app
• postgres-db
• aro-ml-worker
• prometheus
• grafana

All with status: Up ✅

Access Dashboards

Service	URL	Default Credentials
Grafana	http://<your-vm-ip>:3000	admin / admin
Prometheus	http://<your-vm-ip>:9090	—

---

Test the Full System Loop

1. Start Services

./runAro.sh

2. Post Metric Data

# Normal Data
curl -X POST -H "Content-Type: application/json" \
  -d '{"source": "vm-prod-01", "metricType": "CPU_USAGE", "value": 0.85}' \
  http://<your-vm-ip>:8080/api/metrics

# Anomaly Data
curl -X POST -H "Content-Type: application/json" \
  -d '{"source": "vm-prod-01", "metricType": "CPU_USAGE", "value": 25.0}' \
  http://<your-vm-ip>:8080/api/metrics

3. Observe Anomaly Detection

Monitor the ML worker logs:

sudo podman run --name aro-ml-worker-run --replace --network private-net aro-ml-worker:latest
sudo podman logs -f aro-ml-worker

Within 60 seconds, you should see log entries highlighting the anomaly detection event.