Towards a scenario-based solution for extreme metocean event simulation applying urgent computing

摘要

AbstractToday, metocean investigations, combined with forecasts and analysis of extreme events, require new design and development approaches because of their complexity. Extreme metocean events forecasting and prevention is an urgent computing task from decision-making and for reaction point of view. In this case, urgent computing scenario is an essential part that should be included in the hazard simulation and prevention system However, existed urgent computing technological concepts does not perfectly fit all tasks in a frame of extreme metocean events simulation. Many of these tasks should be executed during the overall lifecycle of hazard prevention system that includes not only urgent scenario but research part, as well In this paper, we decompose all tasks in three groups by most significant computational aspects (taking into consideration different criteria of data processing and high-performance contributions) and suggest a new solution that is adaptable for both research in normal (non-urgent) and urgent computing modes, where potential tasks can be structured in the form of scenarios. Suggested solution implements CLAVIRE platform core and extends its with advanced features (regarding simulation frequency, computational performance and data-driven computing). As an example in the metocean subject area, a complex application for Baltic Sea simulations is presented. The case studies describe three scenarios with proposed infrastructure features that are the most interesting for highlighting relevant problems of metocean simulations within the Baltic Sea. These features are: computational optimization possibilities for real-time forecast system calibration; data replacement capabilities within retrospective ensemble extreme values analysis; and hard deadline features within uncertainty analysis of an urgent scenario for complex floods.Highlights•Scientific tasks of metocean forecasting that demand new approaches to computational infrastructure.•Hazard reaction time limits lead us to developing a scenario-based solution with applying of urgent computing.•Platform solution for extreme metocean events should contain all advanced technological features of existed distributed computational systems and efficiently fulfill complex research.•Platform should be oriented for extended scenarios against basic forecast scenario.