Document Type
Article
Publication Date
2022
DOI
10.4236/jcc.2022.109005
Publication Title
Journal of Computer and Communications
Volume
10
Issue
9
Pages
66-80
Abstract
To improve the power consumption of parallel applications at the runtime, modern processors provide frequency scaling and power limiting capabilities. In this work, a runtime strategy is proposed to maximize performance under a given power budget by distributing the available power according to the relative GPU utilization. Time series forecasting methods were used to develop workload prediction models that provide accurate prediction of GPU utilization during application execution. Experiments were performed on a multi-GPU computing platform DGX-1 equipped with eight NVIDIA V100 GPUs used for quantum chemistry calculations in the GAMESS package. For a limited power budget, the proposed strategy may deliver as much as hundred times better GAMESS performance than that obtained when the power is distributed equally among all the GPUs.
Original Publication Citation
Sosonkina, M., Sundriyal, V., & Vallejo, J. L. G. (2022). Runtime power allocation based on multi-GPU utilization in GAMESS. Journal of Computer and Communications, 10(9), 66-80. https://doi.org/10.4236/jcc.2022.109005
Repository Citation
Sosonkina, Masha; Sundriyal, Vaibhav; and Galvez Vallejo, Jorge Luis, "Runtime Power Allocation Based on Multi-GPU utilization in GAMESS" (2022). Electrical & Computer Engineering Faculty Publications. 337.
https://digitalcommons.odu.edu/ece_fac_pubs/337
Comments
Copyright © 2022 by authors and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution 4.0 International License (CC BY 4.0).