IRIS is an intelligent runtime system for extremely heterogeneous architectures. IRIS discovers available functionality, manage multiple diverse programming systems (e.g., OpenMP, CUDA, HIP, Level Zero, OpenCL, Hexagon) simultaneously in the same application, represents data dependencies, orchestrates data movement proactively, and allows configurable work schedulers for diverse heterogeneous devices.
https://iris-programming.github.io
If you use IRIS, please cite the following paper:
Jungwon Kim, Seyong Lee, Beau Johnston, and Jeffrey S. Vetter. 2021. IRIS: A Portable Runtime System Exploiting Multiple Heterogeneous Programming Systems. In Proceedings of the 25th IEEE High Performance Extreme Computing Conference (HPEC ‘21). 1–8.
@inproceedings{Kim:2021:IRIS,
author={Jungwon Kim, Seyong Lee, Beau Johnston, and Jeffrey S. Vetter},
title={IRIS: A Portable Runtime System Exploiting Multiple Heterogeneous Programming Systems},
booktitle={Proceedings of the 25th IEEE High Performance Extreme Computing Conference},
series={HPEC '21},
year={2021},
pages={1-8},
doi={}
}
If you would like more information on IRIS, please either submit an issue at https://github.com/ORNL/iris/issues, start a discussion at https://github.com/ORNL/iris/discussions, or email vetter@computer.org.