%0 Journal Article %T An Effective Git And Org-Mode Based Workflow For Reproducible Research %+ Middleware efficiently scalable (MESCAL) %+ Laboratoire d'Informatique de Grenoble (LIG) %+ PrograMming and scheduling design fOr Applications in Interactive Simulation (MOAIS) %A Stanisic, Luka %A Legrand, Arnaud %A Danjean, Vincent %< avec comité de lecture %@ 0163-5980 %J Operating Systems Review %I Association for Computing Machinery %V 49 %P 61 - 70 %8 2015 %D 2015 %R 10.1145/2723872.2723881 %K Starpu-simgrid %K Reproducible Research %K Git %K Org-mode %Z Computer Science [cs]/Distributed, Parallel, and Cluster Computing [cs.DC]Journal articles %X In this paper we address the question of developing a lightweight and effective workflow for conducting experimen-tal research on modern parallel computer systems in a repro-ducible way. Our approach builds on two well-known tools (Git and Org-mode) and enables to address, at least par-tially, issues such as running experiments, provenance track-ing, experimental setup reconstruction or replicable analysis. We have been using such a methodology for two years now and it enabled us to recently publish a fully reproducible article [12]. To fully demonstrate the effectiveness of our proposal, we have opened our two year laboratory notebook with all the attached experimental data. This notebook and the underlying Git revision control system enable to illus-trate and to better understand the workflow we used. %G English %2 https://inria.hal.science/hal-01112795/document %2 https://inria.hal.science/hal-01112795/file/SIGOPS_paper.pdf %L hal-01112795 %U https://inria.hal.science/hal-01112795 %~ UGA %~ CNRS %~ INRIA %~ UNIV-GRENOBLE1 %~ UNIV-PMF_GRENOBLE %~ INPG %~ INRIA-RHA %~ LIG %~ INRIA_TEST %~ LORIA2 %~ TESTALAIN1 %~ INRIA2 %~ INRIA-RENGRE %~ POLYTECH-GRENOBLE %~ ANR %~ LIG_SIDCH