This paper addresses non-linear gyrokinetic simulations of ion temperature gradient (ITG) turbulence in tokamak plasmas. The electrostatic Gysela code is one of the few international 5D gyrokinetic codes able to perform global, full-f and flux-driven simulations. Its has also the numerical originality of being based on a semi-Lagrangian (SL) method. This reference paper for the Gysela code presents a complete description of its multi-ion species version including: (i) numerical scheme, (ii) high level of parallelism up to 500k cores and (iii) conservation law properties.

BackgroundOsteogenesis imperfecta (OI), commonly called “brittle bone disease”, is a genetic disorder characterised by increased bone fragility and decreased bone density due to quantitative and/or qualitative abnormalities of type I collagen. Different types of OI exist, from mild to severe; they may lead to death, multiple bone fractures, skeletal deformity and short stature.MethodsSevere cases are usually diagnosed before birth and may incite the parents to choose therapeutic abortion, whereas milder cases are much more difficult to diagnose and may be sometimes confused with non-accidental injury (NAI) (“child abuse”) in young children. Whatever the degree of severity, conventional radiography still remains the mainstay in diagnosing OI.ResultsThe prognosis of this disorder has changed in the last few years thanks to biphosphonate therapy.ConclusionThe aim of this pictorial review is to illustrate the radiographic manifestations of OI, including in children receiving biphosphonates, and to outline specific patterns that help differentiate OI from NAI when necessary.Key Points• The main radiographic features of OI are osteopenia, bone fractures and bone deformities.• Some radiographic features depend on the type of OI or may be encountered with biphosphonates.

Résumé. Les simulations gyrocinétiques ont des coûts en calcul extrêmement importants. Jusqu'à maitenant, le code semi-Lagrangien Gysela réalisait de grandes simulations en employant quelques milliers de coeurs (8k coeurs typiquement). Il est prévu que des simulationsà grain plus fin et incluant lesélectrons cinétiques augmentent ces besoins d'un facteur important, fournissant un exemple d'application nécessitant des machines Exascale. Ce papier présente notre travail pour améliorer Gysela afin de viser une architecture qui offre une direction possible vers l'Exascale : la Blue Gene/Q. Après avoir analysé les limitations du code sur cette architecture, nous avons mis en oeuvre trois types d'améliorations : des améliorations de performances de calcul, des améliorations de consomation mémoire et des améliorations d'E/S disque. Nous montrons que suiteà ces travaux, le code monte en charge au delà de 32k coeurs avec des performances bien améliorées. Il devient ainsi possible de viser les machines les plus performantes disponibles et de gérer des cas physiques nettement plus grands.Abstract. Gyrokinetic simulations lead to huge computational needs. Up to now, the semiLagrangian code Gysela performed large simulations using a few thousands cores (8k cores typically). Simulation with finer resolutions and with kinetic electrons are expected to increase those needs by a huge factor, providing a good example of applications requiring Exascale machines. This paper presents our work to improve Gysela in order to target an architecture that presents one possible way towards Exascale: the Blue Gene/Q. After analyzing the limitations of the code on this architecture, we have implemented three kinds of improvement: computational performance improvements, memory consumption improvements and disk i/o improvements. As a result, we show that the code now scales beyond 32k cores with much improved performances. This will make it possible to target the most powerful machines available and thus handle much larger physical cases.