Publications

Peer-reviewed papers on how SCALE compiles CUDA for GPUs from more than one vendor, and how it performs.

Scaling the Extended OpenDwarfs: Evaluating Cross-Vendor CUDA Performance Portability with SCALE
Beau Johnston, Chris Kitching, Matthew Ireland, Michael Søndergaard
ICPP Workshops ’26, 55th International Conference on Parallel Processing, 2026
Read the paper
Cross-Vendor GPU Programming: Extending CUDA Beyond NVIDIA
Manos Pavlidakis, Chris Kitching, Nicholas Tomlinson, Michael Søndergaard
HCDS ’25, 4th Workshop on Heterogeneous Composable and Disaggregated Systems, 2025
Read the paper
SCALE: Ahead-of-Time Compilation of CUDA for AMD GPUs
Manos Pavlidakis, Chris Kitching, Nicholas Tomlinson, Michael Søndergaard
Middleware ’24, 25th International Middleware Conference (Demos and Posters), 2024
Read the paper

Scaling the Extended OpenDwarfs: Evaluating Cross-Vendor CUDA Performance Portability with SCALE