UPCommons. Portal del coneixement obert de la UPC
Publicación
Contribución de congreso (ConferenceOutput)
DPU Offloading Programming with the OpenMP API
Data processing units (DPUs) as network co-processors are an emerging trend in our community, with plenty of opportunities yet to be explored. These have been generally used as domain-specific accelerators transparent to application developers; In the HPC field, DPUs have been used as MPI accelerators, but also to offload some tasks from the general-purpose processor. However, the latter required application developers to deploy MPI ranks in the DPUs, as if they were remote (weak) compute nodes, hence considerably hindering programmability. The wide adoption of OpenMP as the threading model in the HPC arena, along with that of GPU accelerators, is making OpenMP offloading to GPUs a wide trend for HPC applications. In this paper we introduce, for the first time in the literature, OpenMP offloading support for network co-processor DPUs. We present our design in LLVM to support OpenMP standard offloading semantics and discuss the programming productivity advantages with respect to the existing MPI-based programming model. We also provide the corresponding performance analysis demonstrating competitive results in comparison with the MPI baseline., The authors would like to express their sincere gratitude to Gilad Shainer, Richard Graham, Gil Bloch, and Yong Qin for
their support, insightful comments, and suggestions. The research producing this paper received funding from NVIDIA
and the Ministerio de Ciencia e Innovación—Agencia Estatal de Investigación (PCI2021-121958). The authors are grateful
for the support from the Department of Research and Universities of the Government of Catalonia to the AccMem
(Code: 2021 SGR 00807). Antonio J. Peña was partially supported by the Ramón y Cajal fellowship RYC2020-030054-I
funded by MCIN/AEI/10.13039/501100011033 and, by “ESF Investing in your future”, Peer Reviewed