Discussions:New Publication Model

From cTuning.org

Revision as of 13:21, 9 January 2013

New publication model for computer engineering focusing on reproducibility and sharing of experimental results

With the background in physics, we found it extremely disappointing that reproducibility, sharing and statistical mindfulness of results in computer engineering is rarely considered. In fact, it is often simply impossible due to lack of common repositories and tools. Therefore, since 2009, we started using cTuning technology and repository to promote a new publication model for computer engineering (or any other science) which favors sharing of data, models, tools and interfaces for validation and reproducibility by the community. This topic has been accepted for the HiPEAC3 network of excellence (2012-2015) and we are currently building a community around this model. We plan to have a community meeting to discuss practical implementations in October 2012 in Ghent (Belgium). If you are interested to join this collaborative effort, please contact Grigori Fursin or share your ideas here.

As an example, since 1999, we have been releasing all the tools used for our publications as open-source and we are releasing experimental data through cTuning repository in most of our publications since 2008. Since 2005, we also make our cTuning-related lectures available on-line. This model resulted in multiple collaborative projects to improve predictive models and tools to design and optimize computer systems together with IBM (Israel), Google (USA), ICT (China), University of Edinburgh (UK), UPC (Spain), CAPS Entreprise (France), ISP RAS (Russia), Intel (Illinois), Ghent University (Belgium), UVSQ (France), NCAR (USA), ARC/Synopsys (UK) and others.

Current work in progress

• Establishing a conference or a journal based on cTuning technology where PC members can reproduce and validate results
• Creating a common framework and repository (Collective Mind) for sharing and resuing data, tools and models