The EUPEX researchers share their results and lessons learnt!

We are organising monthly webinars on the results of EUPEX (full programme here).Our eleventh webinar will focus on Cybeletech’s agronomic modelling workflow and how it was ported and optimised on the EUPEX ARM-based platform to run large-scale climate and crop simulations.
Abstract:
How will climate change affect where and when we can grow our crops? Cybeletech builds mathematical models of plant biology to forecast crop development, such as flowering and harvest dates, from soil, weather and crop management data. Within EUPEX, the team ported its simulation workflow, combining C++ crop models and a Python pipeline, to the ARM-based pilot system. They then used it to run hundreds of thousands of simulations combining ECMWF climate scenarios, years, varieties, sowing dates and locations across France and Europe. In this webinar, the speakers will present the agronomic use case and the optimisations that unlocked full utilisation of the node, moving from inter-process communication to Pybind11 and then to free-threaded Python. They will also demonstrate the impact of climate change on maize cultivation up to 2050 through a live demonstrator.
Speakers
Christophe Shaw – CEO, Cybeletech
Christophe Shaw is CEO of Cybeletech, a French AgTech company that turns plant biology into mathematical models for agricultural prediction and optimisation. In the past, he held senior engineering leadership roles at Microsoft.
Margot Simon – Agronomist and Data Scientist, Cybeletech
Margot Simon is an agronomist and data scientist at Cybeletech, where she works on crop modelling and simulation pipelines. Within EUPEX, she ported and benchmarked Cybeletech’s simulation workflow on the SDV.