Workshop:P3HPC: 3rd International Workshop on Performance Portability and Productivity
Authors: Rob Neely (Lawrence Livermore National Laboratory), John Pennycook (Intel Corporation), CJ Newburn (Nvidia Corporation), and Scott Parker (Argonne National Laboratory (ANL))
Abstract: The ability for applications to achieve portability and high performance across computer architectures without sacrificing developer productivity is an open challenge, although success stories are emerging. The increasing popularity of GPU-accelerated systems alongside traditional CPU-based systems has brought this issue front-and-center within the HPC application community, and the trends in future hardware point toward ever-increasing hardware diversity, heterogeneity and programmer complexity. There are multiple approaches being developed and in regular use aimed at managing this complexity through language extensions and abstractions, pragma-based standards, new HPC languages and even new algorithms. General architectural trends can be exploited through common programming techniques, language support and robust standards as the community experience matures. The purpose of this workshop is to provide an opportunity for attendees to share ideas, practical experiences and methodologies for tackling the challenge of achieving performance portability and developer productivity across current and future homogeneous and heterogeneous computer architectures.