Generic Topics - Scientific data
The rapid and steady progress in scientific observation instruments (e.g. Sensors) and simulation tools (which promote in silico experimentation) lead to an explosion of data, in terms of volume and complexity. In the field of climate modeling, for example, the increase in data is such that it should produce hundreds of exabytes of data by 2020. Scientific data are also very complex because of their multiphysics or multiscale nature, with up to hundreds of attributes or dimensions. Handling such data is a big challenge for researchers in data management and analysis, because the techniques are complex and need to go on a massive scale. It is also a major issue for the progress of environmental and life sciences, which produce huge amounts of data that must be accessible, organized, analyzed and shared. This major issue today represents a large international community of scientists from different disciplines and researchers in data management. Teams involved in this axis have both extensive expertise in data management and a long history of collaboration with the local scientific community (agronomy, environment, health). We aim to perform ambitious and original research that addresses, at least, the following aspects: uncertain data management and analysis; knowledge discovery and data mining; heterogeneous data integration and visualization; metadata management by a semantic approach, based on ontologies; online data processing and analysis, including leveraging parallel computing; workflow management in "High Performance Computing" environments; sensitive data security and confidentiality by methods and tools for robust encryption with software and hardware support.