Characterizing Energy Consumption of Third-Party API Libraries using API Utilization Profiles (ESEM'2020 Dataset)

Motivation

This repository contains the data-set used as a basis for our ESEM'2020 paper Characterizing Energy Consumption of Third-Party API Libraries using API Utilization Profiles The dataset is comprised of 2 test scenarios: The first one is based on the paper from Rocha et al. (2019), the second one is based on the commonly known Google Gson library.

Description of the dataset

The dataset is stored in a file *.csv and contains the following data:

• id - an individual identifier
• name - the name of the library examined
• className - the class name as an abbreviation
• method - the name of the executed method
• duration - duration of method execution
• energyConsumption - computed energy consumption
• watts - recorded wattage
• uApi - the computed uAPI profile value

Besides the data, this repository also contains the result images from the ESEM'2020 paper in pdf-file format.

Android I/O Experiment

For the Android I/O Experiment we examined the correlation between API utilization and energy consumption. For the dataset, we took inspiration from Rocha et al. (2019). The dataset uses abbreviations for the examined classes which are further described in the readme.md file

• Filename: io_test_052020.csv

Google Gson Experiment

The Google Gson dataset contains recordings for Google Gson version 2.8.5. Again, the dataset was used as a foundation for examining the correlation between the computed API profiles and energy consumption.

• Filename: gson_test_052020.csv

License

Creative Commons CC-BY