Application-Specific Power Consumption Monitor

This program measures the power consumption of a specific application on macOS by correlating the application's CPU and GPU usage with system power metrics.

Features

• Monitors power consumption of a specific application by name or process ID
• Provides real-time power consumption estimates
• Calculates average and maximum power consumption
• Saves detailed monitoring data to a JSON file for further analysis
• Works on macOS with Apple Silicon or Intel processors

Requirements

• macOS (tested on macOS 24.1.0)
• Python 3.6 or higher
• psutil Python package
• powermetrics command-line tool (built into macOS)

Installation

1. Clone or download this repository
2. Create and activate a virtual environment (recommended):

# Create a virtual environment
python3 -m venv venv

# Activate the virtual environment
source venv/bin/activate

3. Install the required Python packages:

pip install -r requirements.txt

Usage

The program requires sudo privileges to access the powermetrics tool. Make sure your virtual environment is activated:

# Activate the virtual environment (if not already activated)
source venv/bin/activate

# Monitor by application name
sudo python3 Power.py -n "Safari"

# Monitor by process ID
sudo python3 Power.py -p 1234

# Specify monitoring duration and interval
sudo python3 Power.py -n "Chrome" -d 300 -i 2

# To view graph 
 sudo python3 Power.py -n "Cursor" -g

Command-line Arguments

• -n, --name: Application name to monitor
• -p, --pid: Process ID to monitor
• -i, --interval: Sampling interval in seconds (default: 1.0)
• -d, --duration: Monitoring duration in seconds (default: 60)

How It Works

This tool uses a combination of techniques to estimate application-specific power consumption:

1. It uses the psutil library to monitor the CPU and memory usage of the target application
2. It uses macOS's built-in powermetrics tool to collect system-wide power metrics
3. It correlates the application's CPU usage with the system's total power consumption to estimate the application's share of power usage
4. When available, it uses process-specific energy information provided by powermetrics

Limitations

• The power consumption estimates are approximations based on CPU usage correlation
• GPU power attribution is simplified and may not be accurate for GPU-intensive applications
• Background system processes may affect the accuracy of measurements
• Requires sudo privileges to access detailed power metrics

Output

The program displays real-time power consumption information and saves detailed data to a JSON file. The JSON file includes:

• Application information (name, PID)
• Monitoring parameters (duration, interval)
• Summary statistics (average power, maximum power, total energy)
• Detailed time-series data for further analysis

License

This project is open source and available under the MIT License.