DOE OSTI · 1846269
Modeling Data Movement Performance on Heterogeneous Architectures
Abstract
The cost of data movement on parallel systems varies greatly with machine architecture, job partition, and nearby jobs. Performance models that accurately capture the cost of data movement provide a tool for analysis, allowing for communication bottlenecks to be pinpointed. Modern heterogeneous architectures yield increased variance in data movement as there are a number of viable paths for inter-GPU communication. In this paper, we present performance models for the various paths of inter-node communication on modern heterogeneous architectures, including the trade-off between GPUDirect communication and copying to CPUs. Furthermore, we present a novel optimization for inter-node communication based on these models, utilizing all available CPU cores per node. Finally, we show associated performance improvements for MPI collective operations.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Bienz, Amanda, Olson, Luke N., Gropp, William D., Lockhart, Shelby. 2021-09-20. Modeling Data Movement Performance on Heterogeneous Architectures. https://doi.org/10.1109/hpec49654.2021.9622742. https://searchnasa.com/record/osti%3A1846269/modeling-data-movement-performance-on-heterogeneous-architectures
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.