Data and Software Preservation for Open Science (DASPOS)

点击翻译按钮获取中文摘要

Scientific data is being collected at a higher rate than ever. A striking example is in elementary particle physics (EPP) experiments, where collaborations of thousands of scientists collect huge amounts of data at the Large Hadron Collider (LHC), but other disciplines have and are continuing to observe similar growth. The complexity and time frame of these experiments is such that the full scientific potential can only be realized when the data remains accessible and analyzable through extended periods. The possibility of successfully meeting this goal for the necessary long term data preservation requires a novel approach that will make the data and necessary software management more solid and survivable. The long term data preservation will become an even more critical issue as present experimental efforts evolve and the Big Data paradigm develops. The initial efforts of the US community to analyze the large volume of LHC data is being satisfied by the Open Science Grid project, designed to facilitate such large and distributed experiments. DASPOS provides an opportunity to continue to study today's analysis over the long term. The DASPOS project incorporates the present state-of-the-art knowledge in working with data in EPP. This project aims to provide a generic technological framework where the basic difficulties are identified and solved with the aim of advancing these goals simultaneously for several disciplines, which should facilitate the emergence of commonalities and standards.The milestones and the work plan are well structured and adapted to present knowledge. Intense communications via workshops is planned and is a natural path for the goal of inclusion of the various disciplines. The intention to document these workshops is a valuable component of this project, as are concrete goals such as prototypes and software challenges.The DASPOS project is not only sound but also timely. The recent dynamics in data preservation and large data management is now reaching several countries and funding agencies. In fact, several projects at national levels are now installed, including for instance, the PREDON project, financed by CNRS-France in 2012 to prepare a multidisciplinary novel approach to big data management. Other similar initiatives are under study in Germany and Italy. It is no question that synergies will emerge at an international scale, and is also clear that DASPOS will play a pioneering and leading role in this context. In conclusion, the impact and merit of the DASPOS proposal is innovative and potentially transformational. This can be an historical opportunity to make a significant advance in the scientific data management in the context of the "Big Data" challenge.