DOE OSTI · 1378602
Implementing High-Performance Geometric Multigrid Solver with Naturally Grained Messages
Abstract
Structured-grid linear solvers often require manually packing and unpacking of communication data to achieve high performance.Orchestrating this process efficiently is challenging, labor-intensive, and potentially error-prone.In this paper, we explore an alternative approach that communicates the data with naturally grained messagesizes without manual packing and unpacking. This approach is the distributed analogue of shared-memory programming, taking advantage of the global addressspace in PGAS languages to provide substantial programming ease. However, its performance may suffer from the large number of small messages. We investigate theruntime support required in the UPC ++ library for this naturally grained version to close the performance gap between the two approaches and attain comparable performance at scale using the High-Performance Geometric Multgrid (HPGMG-FV) benchmark as a driver.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Shan, Hongzhang, Williams, Samuel, Zheng, Yili, Kamil, Amir, Yelick, Katherine. 2015-10-26. Implementing High-Performance Geometric Multigrid Solver with Naturally Grained Messages. https://doi.org/10.1109/pgas.2015.12. https://searchnasa.com/record/osti%3A1378602/implementing-high-performance-geometric-multigrid-solver-with-naturally-grained-messages
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Search NASA by Indiciae: BibTeX · RIS.