Share:


Comparison of GPU and CPU efficiency while solving heat conduction problems

Abstract

Overview of GPU usage while solving different engineering problems, comparison between CPU and GPU computations and overview of the heat conduction problem are provided in this paper. The Jacobi iterative algorithm was implemented by using Python, TensorFlow GPU library and NVIDIA CUDA technology. Numerical experiments were conducted with 6 CPUs and 4 GPUs. The fastest used GPU completed the calculations 19 times faster than the slowest CPU. On average, GPU was from 9 to 11 times faster than CPU. Significant relative speed-up in GPU calculations starts when the matrix contains at least 4002 floating-point numbers.


Article in English.


GPU ir CPU efektyvumo palyginimas sprendžiant šilumos laidumo uždavinius


Santrauka


Šiame straipsnyje apžvelgtas GPU taikymas įvairiems inžineriniams uždaviniams spręsti, palyginti skaičiavimai naudojant CPU ir GPU, aprašytas šilumos laidumo uždavinys. Įgyvendintas Jakobio metodas naudojant „Python“, „TensorFlow GPU“ biblioteką ir NVIDIA CUDA technologijas. Atlikti skaitiniai eksperimentai naudojant šešis CPU ir keturis GPU įtaisus. Greičiausias nagrinėtas GPU įvykdė skaičiavimus 19 kartų greičiau negu lėčiausias CPU. Naudojant GPU, vidutiniškai skaičiavimai buvo atliekami nuo 9 iki 11 kartų greičiau nei su CPU. Didelis santykinis GPU pagreitėjimas vyko, kai lygiagrečiai buvo apdorojama bent 4002 realiųjų skaičių.


Reikšminiai žodžiai: CUDA, GPU, Jakobio metodas, lygiagretieji skaičiavimai, šilumos laidumo uždavinys.

Keyword : CUDA, GPU, Jacobi iterative algorithm, parallel computing, heat conduction problem

How to Cite
Semenenko, J., Kolesau, A., Starikovičius, V., Mackūnas, A., & Šešok, D. (2020). Comparison of GPU and CPU efficiency while solving heat conduction problems. Mokslas – Lietuvos Ateitis / Science – Future of Lithuania, 12. https://doi.org/10.3846/mla.2020.13500
Published in Issue
Nov 24, 2020
Abstract Views
652
PDF Downloads
487
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

References

Amador, G., & Gomes, A. (2012). Linear solvers for stable fluids: GPU vs CPU. https://www.it.ubi.pt/17epcg/Actas/artigos/17epcg_submission_39.pdf

Bohacek, J., Kharicha, A., Ludwig, A., Wu, M., Holzmann, T., & Karimi-Sibaki, E. (2019). A GPU solver for symmetric positive-definite matrices vs. traditional codes. Computers & Mathematics with Applications, 78(9), 2933–2943.
https://doi.org/10.1016/j.camwa.2019.02.034

Crow, F. (1977). Shadow algorithms for computer graphics. ACM SIGGRAPH Computer Graphics, 11(2), 242–248. https://doi.org/10.1145/965141.563901

Fambrini, F., Iano, Y., Caetano, D. G., Rodriguez, A. A. D., Moya, C., Carrara, E., Rangel, A., Cabello, F. C., Zubem, J. V., del val Cura, L. M., Destro Filho, J. B., Campos, J. R., & Saito, J. H. (2018). GPU Cuda JSEG Segmentation Algorithm associated with Deep Learning Classifier for Electrical Network Images Identification. Procedia Computer Science, 126, 557–565. https://doi.org/10.1016/j.procs.2018.07.290

Filonenko, A., Hernández, D. C., & Jo, K.-H. (2018). Fast smoke detection for video surveillance using CUDA. IEEE Transactions on Industrial Informatics, 14(2), 725–733. https://doi.org/10.1109/TII.2017.2757457

Jacobi, C. G. (2009). Über ein leichtes Verfahren, die in der Theorie der Säkularstörungen vorkommenden Gleichungen numerisch aufzulösen. Crelle’s Journal, 1846(30), 51–94. https://doi.org/10.1515/crll.1846.30.51

Kuckuk, S., & Köstler, H. (2018). Whole program generation of massively parallel shallow water equation solvers. In 2018 IEEE International Conference on Cluster Computing (CLUSTER) (pp. 78–87). IEEE.
https://doi.org/10.1109/CLUSTER.2018.00020

Lu, Y., Ramachandra, A. C., Pham, M., Tu, Y.-C., & Cheng, F. (2019). CuDDI: A CUDA-based application for extracting drug-drug interaction related substance terms from PubMed literature. Molecules, 24(6), Article 1081. https://doi.org/10.3390/molecules24061081

Margaris, A., Souravlas, S., & Roumeliotis, M. (2014). Parallel implementations of the Jacobi linear algebraic systems solve [Conference presentation]. Balkan Conference of Informatics (BCI2007), Sofia, Bulgaria.

Warrena, C., Giannopoulos, A., Gray, A., Giannakis, I., Patterson, A., Wetter, L., & Hamrah, A. (2019). A CUDA-based GPU engine for gprMax: Open source FDTD electromagnetic simulation software. Computer Physics Communications, 237, 208–218. https://doi.org/10.1016/j.cpc.2018.11.007

Williams, L. (1978). Casting curved shadows on curved surfaces. ACM SIGGRAPH Computer Graphics, 12(3), 270–274. https://doi.org/10.1145/965139.807402