https://github.com/bharathbk07/sustanability
https://github.com/bharathbk07/sustanability
Last synced: 3 months ago
JSON representation
- Host: GitHub
- URL: https://github.com/bharathbk07/sustanability
- Owner: bharathbk07
- Created: 2025-03-15T19:11:38.000Z (3 months ago)
- Default Branch: main
- Last Pushed: 2025-03-15T19:13:42.000Z (3 months ago)
- Last Synced: 2025-03-15T20:22:00.465Z (3 months ago)
- Language: Python
- Size: 12.7 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Sustainability Metrics Monitoring
This application monitors system sustainability metrics, including CPU, GPU, and RAM power consumption, and estimates carbon emissions. It logs the data to a CSV file and exposes metrics to Prometheus for real-time monitoring.
## Features
- **Power Consumption Monitoring**: Tracks CPU, GPU, and RAM power usage.
- **Carbon Emissions Estimation**: Calculates emissions based on energy consumption and grid carbon factor.
- **Cloud Environment Detection**: Identifies if the system is running on AWS, Azure, or GCP.
- **Location Detection**: Fetches approximate latitude and longitude of the machine.
- **Prometheus Integration**: Exposes metrics for real-time monitoring.
- **CSV Logging**: Optionally logs metrics to a CSV file.
## Requirements
- Python 3.6 or higher
- Prometheus Client Library (`prometheus_client`)
- `psutil` for system monitoring
- `requests` for API calls
- Optional: `pynvml` for NVIDIA GPU monitoring
## Installation
1. Clone the repository:
```bash
git clone
cd Sustainability
```
2. Install dependencies:
```bash
pip install -r requirements.txt
```
3. (Optional) Install `pynvml` for GPU monitoring:
```bash
pip install nvidia-ml-py3
```
## Usage
1. Start the Prometheus HTTP server and begin monitoring:
```bash
python app.py
```
2. Access Prometheus metrics at:
```
http://localhost:9271/metrics
```
3. (Optional) Enable CSV logging by setting `write_to_file = True` in the script.
## Configuration
- **Grid Carbon Factor**: Update the `GRID_CARBON_FACTOR` constant to match your region's carbon intensity (default: 400 gCO2e/kWh).
- **CSV File Name**: Change the `CSV_FILE` constant to specify a different file name for logging.
## Metrics
### Numeric Metrics
The following metrics are exposed to Prometheus:
- `duration`: Monitoring duration (seconds)
- `emissions`: Carbon emissions (kgCO2e)
- `emissions_rate`: Emissions rate (kg/s)
- `cpu_power`: CPU power usage (Watts)
- `gpu_power`: GPU power usage (Watts)
- `ram_power`: RAM power usage (Watts)
- `cpu_energy`: CPU energy consumption (kWh)
- `gpu_energy`: GPU energy consumption (kWh)
- `ram_energy`: RAM energy consumption (kWh)
- `energy_consumed`: Total energy consumption (kWh)
### Text Labels
Metadata labels include:
- `project_name`, `country_name`, `region`, `cloud_provider`, `os`, `cpu_model`, `gpu_model`, etc.
## Stopping the Application
Press `Ctrl+C` to stop monitoring. If GPU monitoring is enabled, the script will automatically shut down NVML.
## Notes
- Ensure the system has internet access for location and cloud provider detection.
- GPU monitoring is only available for NVIDIA GPUs with NVML support.
- Prometheus must be configured to scrape metrics from `http://localhost:9271/metrics`.