Metrics and Design of an Instruction Roofline Model for AMD GPUs
ACM Transactions on Parallel Computing · 31.03.2022 · 9 (1) · 1-14
DOI: 10.1145/3505285 ↗Kurz gefasst
Fachartikel in der Zeitschrift ACM Transactions on Parallel Computing, online erschienen am 31.01.2022. Verfasst von Matthew Leinhauser und 5 weiteren Autor:innen. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 25 Literaturangaben. Gefördert von: Office of Science of the U.S. Department of Energy, National Science Foundation.
Abstract (Originalsprache)
Due to the recent announcement of the Frontier supercomputer, many scientific application developers are working to make their applications compatible with AMD (CPU-GPU) architectures, which means moving away from the traditional CPU and NVIDIA-GPU systems. Due to the current limitations of profiling tools for AMD GPUs, this shift leaves a void in how to measure application performance on AMD GPUs. In this article, we design an instruction roofline model for AMD GPUs using AMD’s ROCProfiler and a benchmarking tool, BabelStream (the HIP implementation), as a way to measure an application’s performance in instructions and memory transactions on new AMD hardware. Specifically, we create instruction roofline models for a case study scientific application, PIConGPU, an open source particle-in-cell simulations application used for plasma and laser-plasma physics on the NVIDIA V100, AMD Radeon Instinct MI60, and AMD Instinct MI100 GPUs. When looking at the performance of multiple kernels of interest in PIConGPU we find that although the AMD MI100 GPU achieves a similar, or better, execution time compared to the NVIDIA V100 GPU, profiling tool differences make comparing performance of these two architectures hard. When looking at execution time, GIPS, and instruction intensity, the AMD MI60 achieves the worst performance out of the three GPUs used in this work.
Förderung
- Office of Science of the U.S. Department of Energy
- National Science Foundation
Zitieren
APA
Leinhauser, M., Widera, R., Bastrakov, S., Debus, A., Bussmann, M., & Chandrasekaran, S. (2022). Metrics and Design of an Instruction Roofline Model for AMD GPUs. ACM Transactions on Parallel Computing, 9(1), 1-14. https://doi.org/10.1145/3505285
Vancouver
Leinhauser M, Widera R, Bastrakov S, Debus A, Bussmann M, Chandrasekaran S. Metrics and Design of an Instruction Roofline Model for AMD GPUs. ACM Transactions on Parallel Computing. 2022;9(1):1-14. doi:10.1145/3505285
BibTeX
@article{Leinhauser2022,
author = {Leinhauser, Matthew and Widera, René and Bastrakov, Sergei and Debus, Alexander and Bussmann, Michael and Chandrasekaran, Sunita},
title = {Metrics and Design of an Instruction Roofline Model for AMD GPUs},
journal = {ACM Transactions on Parallel Computing},
year = {2022},
volume = {9},
number = {1},
pages = {1--14},
doi = {10.1145/3505285}
}
EUVIMED