2.45pm to 3.30pm
Weather and Climate Modelling world-wide is a significant user of computational resources.
Not only do Numerical Weather Prediction Models (NWP) require vast numbers of FLOPS, as scientist try to increase the number of grid points (i.e. resolution and hence accuracy of the modelling), but also large amounts of internal data transfer and storage. FLOPS are free is the current mantra in the NWP world.
This talk will illustrate the challenges for one such project, WISER (Weather climate change Impact Study at Extreme Resolution) and how important data transfer and manipulation has become. The issues facing the Next Generation Weather and Climate codes will be illustrated by reference to current developments in the UK (Gungho project) and other codes (e.g. US MPAS project )which take advantage of massively parallel computers.Speakers: Alan Gadian - University of Leeds
View slides Listen to Audio Watch Video Close