@inproceedings{c95e78a4299b4ae6b97bf66b9c7d82ce,
title = "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.",
keywords = "CUDA-Aware, GPU, GPUDirect, MPI, data movement, performance modeling",
author = "Amanda Bienz and Olson, {Luke N.} and Gropp, {William D.} and Shelby Lockhart",
note = "Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE High Performance Extreme Computing Conference, HPEC 2021 ; Conference date: 20-09-2021 Through 24-09-2021",
year = "2021",
doi = "10.1109/HPEC49654.2021.9622742",
language = "English (US)",
series = "2021 IEEE High Performance Extreme Computing Conference, HPEC 2021",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2021 IEEE High Performance Extreme Computing Conference, HPEC 2021",
address = "United States",
}