IDEFIX: A versatile performance-portable Godunov code for astrophysical flows
Geoffroy Lesur, S. Baghdadi, Gaylor Wafflard-Fernandez, J. Mauxion, C. Robert, Mathias van den Bossche
Abstract
Context. The exascale super-computers becoming available rely on hybrid energy-efficient architectures that involve an accelerator such as a graphics processing unit (GPU). Leveraging the computational power of these machines often means a significant rewrite of the numerical tools each time a new architecture becomes available. Aims. We present I DEFIX , a new code for astrophysical flows that relies on the K OKKOS meta-programming library to guarantee performance portability on a wide variety of architectures while keeping the code as simple as possible to the user. Methods. I DEFIX is based on a Godunov finite-volume method that solves the nonrelativistic hydrodynamical (HD) and magnetohy-drodynamical (MHD) equations on various grid geometries. I DEFIX includes a large choice of solvers and several additional modules (constrained transport, orbital advection, nonideal MHD), allowing users to address complex astrophysical problems. Results. I DEFIX has been successfully tested on Intel and AMD CPUs (up to 131 072 CPU cores on Irene-Rome at TGCC) as well as NVidia and AMD GPUs (up to 1024 GPUs on Adastra at CINES). I DEFIX achieves more than 10 8 cell s −1 in MHD on a single NVidia V100 GPU and 3 × 10 11 cell s −1 on 256 Adastra nodes (1024 GPUs) with 95% parallelization efficiency (compared to single node). For the same problem, I DEFIX is up to six times more energy efficient on GPUs compared to Intel Cascade Lake CPUs. Conclusions. I DEFIX is now a mature exascale-ready open-source code that can be used on a large variety of astrophysical and fluid dynamics applications.