Lunch bytes #5 Making GPU Programming More Portable

views comments

An event hosted by the Sheffield RSE team on 1st April 2021, featuring three short talks + Q&A about making GPU programming more portable:

PyFR: Facilitating Heterogeneous GPU Computing from a Homogeneous Codebase - Freddie Witherden, Assistant Professor at Texas A&M University, developer of PyFR, a CUDA accelerated python CFD library
GPU acceleration of Spiking Neural Networks - James Knight, Research Fellow in Computer Science at The University of Sussex, developer of PyNN/GeNN, a GPU enhanced Neuronal Network simulation environment
Portable performance on CPUs and GPUs without resorting to [redacted] - Tom Deakin, Senior Research Associate at Bristol University, researcher in understanding performance portability of massively parallel simulation codes