pyEnergiBridge

pyEnergiBridge is a Python wrapper that enables the collection of software energy consumption data using EnergiBridge.

Installation

You can install pyEnergiBridge via pip. First, clone the repository:

git clone https://github.com/luiscruz/pyEnergiBridge.git
cd pyEnergiBridge

Then, install the package using pip:

pip install .

Configuration

By default, pyEnergiBridge looks for the EnergiBridge binary in a predefined path. You can customize the path to the EnergiBridge binary using configuration files. The configuration files can be placed in the user's home directory or in the project directory.

1. Home Directory Configuration:

Create a file named .pyenergibridge_config.json in your home directory:

{
    "binary_path": "/custom/path/to/EnergiBridge/home"
}

2. Project Directory Configuration:

Create a file named pyenergibridge_config.json in your project directory:

{
    "binary_path": "/custom/path/to/EnergiBridge/home"
}

pyEnergiBridge will first look for the configuration file in the project directory, and if it is not found, it will look for the configuration file in the home directory. If no configuration files are found, it will check if EnergiBridge is available in the system's PATH. If none of these methods provide a valid path, it will raise a FileNotFoundError.

Usage

Basic Usage

To collect energy consumption data within your Python code, follow these steps:

import time
from pyEnergiBridge.api import EnergiBridgeRunner

runner = EnergiBridgeRunner()
runner.start(results_file="results.csv")

# Perform some task
time.sleep(2)

# Stop the data collection and retrieve results
energy, duration = runner.stop()
print(f"Energy consumption (J): {energy}; Execution time (s): {duration}")

Using Decorator

You can also use a decorator to simplify the process:

import time
from pyEnergiBridge.api import with_energibridge

@with_energibridge(results_file="results.csv")
def my_task():
    # Simulate a task running for a short period of time
    time.sleep(2)
    print("Task completed.")

# Execute the task
my_task()

Make sure to check the "results.csv" file to confirm that the energy consumption data has been stored successfully.

IPython Magic Commands

You can use IPython magic commands to measure energy consumption and execution time for code cells and lines directly within Jupyter notebooks.

Cell Magic

Use the %%measure_energy_time_cell magic command to measure the energy consumption and execution time of an entire cell:

%%measure_energy_time_cell
import time
time.sleep(2)
print("Task completed.")

Line Magic

Use the %measure_energy_time_line magic command to measure the energy consumption and execution time of a single line of code:

%measure_energy_time_line time.sleep(2)
print("Task completed.")

Usage With Docker

Due to different process management in Docker containers, you need to set the is_containerized flag on instantiation of the EnergiBridgeRunner:

from pyEnergiBridge.api import EnergiBridgeRunner

runner = EnergiBridgeRunner(is_containerized=True)

Note: this only works for Linux Docker containers and will fail on Windows.