The International Conference for High Performance Computing, Networking, Storage, and Analysis
The International Conference for High Performance Computing, Networking, Storage, and Analysis
Workshop: Tenth Workshop on Accelerator Programming and Directives (WACCPD 2023)
Authors: Diego A. Roa Perdomo (University of Delaware, CAPSL; Argonne National Laboratory (ANL)); Rodrigo Ceccato and Rémy Neveu (University of Campinas, Argonne National Laboratory); Hervé Yviquel (University of Campinas); Xiaoming Li (University of Delaware); Jose M. Monsalve Diaz (Argonne National Laboratory); and Johannes Doerfert (Lawrence Livermore National Laboratory)
Abstract: We provide an automated framework that utilizes complex hardware links while preserving the simplified abstraction level for the user. Through the decomposition of user-issued memory operations into architecture-aware sub-tasks, we automatically exploit generally underused connections of the system. The operations we support include moving, distribution, and consolidation of memory across the node. For each of them, our Auto-Strategyzer framework proposes a task graph that transparently improves performance, in terms of latency or bandwidth, compared to naive strategies. For our evaluation, we integrated the Auto-Strategyzer as a C++ library into the LLVM-OpenMP runtime infrastructure. We demonstrate that some memory operations can be improved by a factor of 5x compared to naive versions. Integrated into LLVM/OpenMP, our Auto-Strategyzer accelerates cross-device memory movement by a factor of 1.9x, for large transfers, resulting in approx 6% end-to-end execution time decrease for a scientific proxy application.
Back to Tenth Workshop on Accelerator Programming and Directives (WACCPD 2023) Archive Listing