ENVRI PLUS - Interoperable data processing for environmental ri projects: system design.

Candela L, Coro G, Pagano P, Panichi G, Atkinson M, Filgueira R, Bailo D, Enell Cf, Fiebig M, Haslinger F, Hellström M, Vermeulen A, Lankreijer H, Huber R, Joussaume S, Guglielmo F, Mendez V

Data processing and analysis  envri 

Data processing is a very wide area or domain because of a series of characteristics including the contexts resulting from diverse application scenarios, the great variety of processes to be enabled, the large set of enabling technologies and solutions. One of the consequences of this large variety is that each software solution for data processing only manages to address parts, i.e. it is difficult to imagine a single solution that is equally suitable for any (or even most) application scenarios and contexts. This deliverable illustrates that scope and diversity by reporting detailed practices and requirements from seven of the ENVRI RIs. It then describes the design of a single data processing solution that will help meet a substantial range of requirements in a representative range of contexts. That approach is conceived to be (a) suitable for serving the needs of scientists involved in ENVRI RIs, (b) open and extensible both with respect to the algorithms and methods it enables and the computing platforms it relies on to execute those algorithms and methods, (c) open-science-friendly, i.e. it is capable of incorporating every algorithm and method integrated into the data processing framework as well as any computation resulting from the exploitation of integrated algorithms into a "research object" catering for citation, reproducibility, repeatability and provenance.