Layered Reproducibility for High-Performance Computing Applications: The Feel++ and Ktirio Urban Buildings Case Study
The High-Performance Computing (HPC) community struggles with reproducibility due to complex software stacks, heterogeneous hardware, and expensive validation cycles. This contribution presents a layered reproducibility strategy in which Feel++ provides a traceable software supply chain and Ktirio Urban Buildings (KUB) builds on it to manage versioned urban-building data, executable simulations, and checksum-verifiable result artifacts. Feel++ treats continuous integration (CI) plans, CMake presets, packaging manifests, public software metadata, operations commands, containers, software bill of materials (SBOM)/provenance metadata, and Feel++ Benchmarking reports for Feel++ and KUB cases as repositoryowned artifacts. KUB applies this foundation to City Energy Model Database (CEMDB) datasets, manifests, Findable, Accessible, Interoperable, and Reusable (FAIR) packaging, and verifiable outputs for computationally expensive simulations. The case study shifts reproducibility from late manual packaging to an artifact chain linking source commits, packages, containers, benchmark configurations, dataset versions, provenance records, and generated results.