Graphics processing unit resource management for computational fluid dynamics

被引：0

作者：

Weng Y. ^{[1
]}

Zhang X. ^{[1
]}

Lu Y. ^{[1
]}

机构：

[1] School of Computer Science and Engineering, Sun Yat-sen University, Guangzhou

来源：

Guofang Keji Daxue Xuebao/Journal of National University of Defense Technology | 2022年 / 44卷 / 05期

关键词：

computational fluid dynamics; graphics processing unit; resource management; scheduling;

D O I：

10.11887/j.cn.202205004

中图分类号：

学科分类号：

摘要：

Aiming at the problem of low resource utilization of GPU ( graphics processing unit) in the process of solving CFD ( computational fluid dynamics)，a CFD-oriented GPU resource optimization management scheme was proposed. Based on the characterization of the CFD and tasks running concurrently, a reasonable scheduling scheme was designed. By dynamically changing the startup scale and time of different tasks, our method was able to reduce resource competition while improving the effective use of GPU resources. The experimental results show that compared with the baseline method, the average speedup ratio of our proposed resource management scheme reaches 1. 64 x speedup under different task scales, and the use of GPU hardware resources has also been significantly improved. © 2022 National University of Defense Technology. All rights reserved.

引用

页码：35 / 44

页数：9

共 49 条

[11]

SYNYLO K, KRUPKO A, ZAPOROZHETS O, Et al., CFD simulation of exhaust gases jet from aircraft engine, Energy, 213, (2020)

[12]

KAUSHAL P, SHARMA H K., Concept of computational fluid dynamics ( CFD) and its applications in food processing equipment design, Journal of Food Processing and Technology, 3, 1, (2012)

[13]

LIU XC, ZHONG Z M, XUK, A hybrid solution method for CFD applications on GPU-accelerated hybrid HPC platforms [ J ], Future Generation Computer Systems, 56, pp. 759-765, (2016)

[14]

SLOTNICK J, KHODADOUST A, ALONSO J, Et al., CFD vision 2030 study: a path to revolutionary computational aerosciences [ R ], (2014)

[15]

CUDA [ CP/OL ]

[16]

Nvprof

[17]

Rocprofiler

[18]

KAMBADUR M, HONG S, CABRAL J, Et al., Fast computational GPU design with GT-pin [ C ], Proceedings of IEEE International Symposium on Workload Characterization, pp. 76-86, (2015)

[19]

VILLA O, STEPHENSON M, NELLANS D, Et al., NVBit: a dynamic binary instrumentation framework for NVIDIA GPUs, Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture, pp. 372-383, (2019)

[20]

ZHANG X W, SHCHERBAKOV E., DELTA: validate GPU memory profiling with microbenchmarks [ C ], Proceedings of the International Symposium on Memory Systems, pp. 97-104, (2020)

← 1 2 3 4 5 →