finopsmetrics 0.3.1

FinOps Metrics Platform - Free Community Edition for AI/ML Cost Observability

FinOpsMetrics - Community Edition

Free, Proprietary FinOps Platform for AI/ML Cost Observability

FinOpsMetrics Community Edition is a powerful, free-to-use platform for real-time cost monitoring, tracking, and optimization across multi-cloud environments and AI/ML workloads.

🎯 Key Features

Multi-Cloud Cost Observability

• AWS Integration: EC2, EKS, Lambda, RDS, S3, CloudWatch, Cost Explorer
• Azure Integration: VMs, AKS, Functions, SQL Database, Monitor, Cost Management
• GCP Integration: Compute Engine, GKE, Cloud Functions, Cloud SQL, Monitoring, Billing
• Real-time telemetry collection via agent-based architecture

AI/ML Workload Monitoring

• LLM training cost tracking
• RAG pipeline observability
• Agent workflow metrics
• GPU utilization monitoring
• Model training performance analytics

Infrastructure as Code (IaC) Cost Analysis

• Terraform cost estimation
• CloudFormation analysis
• ARM template evaluation
• Pre-deployment cost forecasting

SaaS & Third-Party Service Tracking

• OpenAI API cost monitoring
• Anthropic Claude API tracking
• Custom API integrations
• Multi-provider cost aggregation

Advanced Cost Management

• ML-powered anomaly detection
• Tag-based cost attribution
• Budget policies and alerts
• Cost allocation and chargeback
• Custom plugin framework

VizlyChart Visualization Library

• 15+ chart types (line, bar, scatter, candlestick, radar, heatmap, treemap, etc.)
• Real-time data streaming
• Interactive dashboards
• Financial time-series charts
• GPU-accelerated rendering (optional)
• Export to PNG, SVG, HTML

📦 Installation

Basic Installation

pip install finopsmetrics

With Cloud Provider Support

# AWS
pip install finopsmetrics[aws]

# Azure
pip install finopsmetrics[azure]

# GCP
pip install finopsmetrics[gcp]

# All cloud providers
pip install finopsmetrics[all]

Development Installation

git clone <your-private-repo>
cd finopsmetrics-community
pip install -e ".[dev]"

🚀 Quick Start

1. Start the Observability Server

finopsmetrics-server --host 127.0.0.1 --port 8080

2. Deploy Telemetry Agents

from agents.aws_telemetry_agent import AWSTelemetryAgent

# AWS agent automatically discovers resources and calculates costs
agent = AWSTelemetryAgent(
    openfinops_endpoint="http://localhost:8080",
    aws_region="us-west-2"
)
agent.run_continuous(interval_seconds=300)

3. Query Cost Data

from finopsmetrics.observability import CostObservatory

cost_obs = CostObservatory()
summary = cost_obs.get_cost_summary(time_range_hours=24)

print(f"Total cost (24h): ${summary['total_cost_usd']:.2f}")
print(f"Top service: {summary['top_services'][0]}")

4. Create Visualizations

from finopsmetrics.vizlychart import LineChart

chart = LineChart(
    title="Daily Cloud Costs",
    x_label="Date",
    y_label="Cost ($)"
)
chart.add_series(dates, costs, label="AWS", color="blue")
chart.render()
chart.save("costs.png")

🎨 VizlyChart Examples

# Financial candlestick chart
from finopsmetrics.vizlychart import CandlestickChart

chart = CandlestickChart(title="Cost Trends")
chart.add_data(dates, opens, highs, lows, closes)
chart.render()

# Heatmap for resource utilization
from finopsmetrics.vizlychart import HeatmapChart

chart = HeatmapChart(title="CPU Utilization")
chart.add_data(matrix_data, x_labels=hours, y_labels=servers)
chart.render()

# Real-time streaming chart
from finopsmetrics.vizlychart import RealtimeChart

chart = RealtimeChart(title="Live Metrics", buffer_size=100)
chart.start_stream(data_source, update_interval=1.0)

🏗️ Architecture

FinOpsMetrics uses an agent-based architecture:

1. Telemetry Agents (separate processes) deployed in cloud accounts
2. Agents automatically discover resources via cloud provider APIs
3. Agents query metrics (CPU, memory, network) and calculate costs
4. Agents send data to ObservabilityHub via REST API
5. ObservabilityHub receives and processes telemetry
6. CostObservatory aggregates costs, tracks budgets, triggers alerts
7. Dashboards display real-time metrics and visualizations

📊 Available Commands

# Start observability server
finopsmetrics-server --port 8080

# View CLI help
finopsmetrics --help

# Run with custom configuration
finopsmetrics-server --config /path/to/config.yaml

🔌 Plugin System

Create custom plugins to extend functionality:

from finopsmetrics.plugins import BasePlugin

class CustomCostPlugin(BasePlugin):
    def process_cost_entry(self, entry):
        # Custom cost processing logic
        return modified_entry

    def generate_report(self):
        # Custom reporting
        pass

Register plugins in config.yaml:

plugins:
  - module: my_plugins.custom_cost
    class: CustomCostPlugin
    enabled: true

🔐 License

Proprietary - Free Community Edition

FinOpsMetrics Community Edition is free to use for personal and commercial purposes under the following conditions:

✅ Permitted:

• Install and use for any lawful purpose
• Use in production environments
• Commercial use without fees
• Evaluation and testing

❌ Restricted:

• No reverse engineering or decompilation
• No redistribution, sale, or sublicense
• No derivative works
• No use to build competing products
• No source code access beyond distributed form

See LICENSE for complete terms.

🚀 Upgrade to Enterprise

For advanced features, upgrade to FinOpsMetrics Enterprise:

• AI-powered cost optimization recommendations
• Advanced analytics and forecasting
• Workspace collaboration features
• Version control for budgets and policies
• GIS-based regional analysis
• GPU-accelerated processing
• Priority support

Contact: [e-mail]

📚 Documentation

• Installation Guide: See above
• API Reference: Coming soon
• Telemetry Agent Setup: See agents/README.md
• VizlyChart Gallery: See examples/vizlychart/

🤝 Support

• Issues: Contact [e-mail]
• Commercial Licensing: [e-mail]
• Enterprise Sales: [e-mail]

📈 Roadmap

• Kubernetes cost attribution
• Enhanced ML model training tracking
• GraphQL API
• Mobile dashboard app
• Advanced FinOps best practice recommendations

---

FinOpsMetrics - Intelligent Cost Observability for Cloud and AI/ML Workloads

Copyright © 2025 Infinidatum. All rights reserved.