The first annual state-of-the-art review from work package 3, which builds the code analysis, quality assurance and artefact generation parts of the GreenCode pipeline. It reviews five areas and, for each, names the gaps GreenCode intends to fill: mapping a software system's codebase (with a parser-driven, model-driven reverse engineering approach based on ANTLR4); static code analysis with a focus on energy, comparing formal verification, model checking and abstract interpretation and their scalability; benchmark analysis and the lack of accepted benchmarks for energy prediction methods; quality assessment of generative-AI outcomes; and static analysers for compiled software. The conclusion identifies the project's key opportunity as combining effect handlers with abstract interpretation so that resource and energy usage can be analysed, traced and even enforced statically within application code, and includes a systematic literature review of approaches for predicting software energy consumption through static analysis.
Authors: Andreas Jedlitschka (ed.), Daniel Esteban Villamil (ed.), Patricia Kelbert, Michele Albano, Chris Dean, Asger Posborg Jæhger, Adam Trendowicz, Anna Maria Vollmer, Julien Siebert, Sven Theobald, Violeta Bonet Vila, Daniel Carlos Do Vale Ramos, Pedro Faria, Rita Costa, Zita Vale, Amir Soltani, Alireza Mehri, Gunel Jahangirova, Jie Zhang, Mohammad Reza Mousavi, Hugo Araujo, Marcos Checa, Cosme Gonzalez, Christian Koerner, Ezgi Sarikaya, Matthias Saft, Angel Cataron, Fridtjof Siebert, Michael Lill, Juanmiguel Gomez, Andrea Pabon, Haluk Gokmen, Osman Çaylı, Sinan Kılıç, Yücel Şentürk, Constantin Deneke, Johannes Passand, Mario Paja, Steffen Ruesch, Stephan Rediske
Contributing partners: Fraunhofer IESE, Universidad Carlos III de Madrid, VBT, ZAL, University of Augsburg, Trier University of Applied Sciences, Tokiwa Software GmbH, Siemens
