User-friendly programming of GPU-enhanced clusters

By developing a simple directive-based programming model and its accompanying fully automated source-to-source code translator and domain-specific optimizer, we aim to greatly simplify the task of programming scientific codes that can run efficiently on accelerator-enhanced computer clusters. This project is motivated by an urgent need from the community of computational scientists for programming methodologies that are easy to use, while capable of harnessing especially the non-conventional computing resources, such as GPUs, that dominate today's HPC field. Based on a proof-of-concept work that has already successfully automated C-to-CUDA translation and optimization restricted to the single-GPU scenario and stencil methods, the proposed project aims to greatly enhance the success by extending to the following topics:

  1. improving the newly developed directive-based programming model and its accompanying framework of automated code translation and optimization
  2. extending to the scenario of multiple GPUs
  3. extending to the scenario of GPU-accelerated CPU clusters
  4. tackling a number of real-world scientific codes

The project has the potential of considerably enhancing the productivity of computational scientists, to let them focus more on their scientific investigations at hand, instead of spending precious time on painstakingly writing complex codes.

Funding source:

Research Council of Norway, FRINATEK program

All partners:

  • Simula Research Laboratory
  • University of California, San Diego (UCSD)
  • San Diego Supercomputer Center (SDSC)
  • National University of Defense Technology (NUDT)

Find publication


Year published


High Performance Computing