Energy-Aware Software Design: Experimental Setup, Data, and Results
Description
Motivated by data transparency and the goal of shortening the length of our paper, the data utilized during the development and experimental evaluation of our algorithm is shared here. The content in this page includes the following: - The experimental setup for every sample in test cases 1 through 4, and the setup of the final experiment with Blender. - The RADIANCE+ project (RADIANCE+ project.rad) with all the data pertaining to the the setup of the BBCP profiles, the expert algorithm, and the complimentary data involved in each sample. Consult the "Steps to reproduce" section available below for detailed information about this file. - The results per sample for each test case, which are replicable through the RADIANCE+ project. RADIANCE+ live is accessible through the following URL: https://joalago.github.io/RADIANCE_latest/ More information about RADIANCE+ can be found in its official webpage: https://webdiis.unizar.es/~silarri/prot/RADIANCEplus/ RADIANCE+ implements the BBCP DSL, which we used during the development of the profiles employed by our algorithm. For more information about the BBCP, you can access its official webpage here: https://webdiis.unizar.es/~silarri/prot/BBCP/ More information about our methodology for green software design can be consulted here (embargoed until 2025-M07-D02): https://zaguan.unizar.es/record/160828
Files
Steps to reproduce
The RADIANCE+ project (RADIANCE+ project.rad) must be imported into the live version of our latest distribution of RADIANCE+ to reproduce our results: https://joalago.github.io/RADIANCE_latest/ The GitHub repository will be publicly published once the embargo mentioned above is lifted. To load the project: 1. Click on the "replace project" button located in the landing page. 2. Select "complete database" in the modal that opens after step 1. 3. Drop the project (RADIANCE+ project.rad) downloaded from this page in the dropzone. 4. In the top navigation bar, click on "Model Builder" and select any of the models available. There is no need to install, re-configure, tinker with or modify any variables to explore the sample profiles in the project and replicate our experiments. If more help is required for using the model builder, a guided tour of it is available in a legacy version of RADIANCE+, RADIANCE, accessible through the following URL: https://joalago.github.io/RADIANCE-demo Note: The RADIANCE+ project provided here won't work in the legacy version used for the guided tour, nor should it be used in an attempt to replicate the results included in this page.
Categories
Funders
- MICIU/AEI/10.13039/501100011033 (Spanish State Research Agency)Grant ID: PID2020-113037RB-I00
- Departamento de Ciencia, Universidad y Sociedad del Conocimiento del Gobierno de Aragón (Government of Aragon)Grant ID: Group Reference T64_23R, COSMOS research group