Environmental Computing 1.0: The Dawn of a concept

Abstract

There is a long history of numerical modelling of various natural phenomena for purposes such as weather prediction or analysis of different earthquake-scenarios. In this paper we present the next logical step: combining multiple models together in a dynamically extensible framework in order to gain a better understanding of the nature and impact of inherently interlinked and dependent environmental phenomena. We call this approach Environmental Computing, which encompasses both the link to a broad range of environmental issues that can be approached using the framework model, and the notion that the component models and their features can be evaluated algorithmically to evaluate the accuracy and applicability of the results in different situations. Reaching this goal requires new technologies, commonly accepted approaches, standards and policies. This multi-pronged approach is necessary, since the combination of different models will bring forth challenges related to the compatibility of the execution environments as well as issues with the syntax and semantics of data. The data challenge applies both to the input of the while model ensemble as well as mechanisms of inter-model data exchange. To showcase the progress made towards this goal so far, we will present the technical and operational frameworks for the environmental multi-modelling, as well as specific case studies that have acted as proofs-of-concepts or pilot tests establishing the state of the art in this domain. These case studies, together with a short summary of related work, will lay the foundations for a discussion on the methods that could be used to assess the impact and benefits of Environmental Computing in this and other contexts.

Wine in the Cloud, or: Smart Vineyards with a Distributed "Extreme Data Database" and Supercomputing

“…European supercomputing centres, in particular in the academic sector, have long recognised the importance of Environmental Computing [3]. The respective applications tend to significantly drive innovation at these centres, posing challenges beyond traditional, simulation-centric HPC, as for example (i) cloud-native computing services have to be offered together with computing and data-flow orchestration, and (ii) the centres must be able to efficiently process huge datasets in database-management systems (DBMS), often used in geoscience (e.g.…”