Proceedings of the Third International Conference on Computing and Wireless Communication Systems, ICCWCS 2019, April 24-25, 2019, Faculty of Sciences, Ibn Tofaïl University -Kénitra- Morocco

Research Article

GPU optimized parallel implementation of NaSch traffic model

Download560 downloads
  • @INPROCEEDINGS{10.4108/eai.24-4-2019.2284233,
        author={Yassine  El Hafid and Abdessamad  El Rharras and Mohamed  Wahbi and Rachid  Saadane},
        title={GPU optimized parallel implementation of NaSch traffic model},
        proceedings={Proceedings of the Third International Conference on Computing and Wireless Communication Systems, ICCWCS 2019, April 24-25, 2019, Faculty of Sciences, Ibn Tofa\~{n}l University -K\^{e}nitra- Morocco},
        publisher={EAI},
        proceedings_a={ICCWCS},
        year={2019},
        month={5},
        keywords={: gpgpu cpu cuda nasch model road traffic parallel computing optimization},
        doi={10.4108/eai.24-4-2019.2284233}
    }
    
  • Yassine El Hafid
    Abdessamad El Rharras
    Mohamed Wahbi
    Rachid Saadane
    Year: 2019
    GPU optimized parallel implementation of NaSch traffic model
    ICCWCS
    EAI
    DOI: 10.4108/eai.24-4-2019.2284233
Yassine El Hafid1,*, Abdessamad El Rharras1, Mohamed Wahbi1, Rachid Saadane1
  • 1: SIRC/LaGeS EHTP
*Contact email: yassine.el.hafid@gmail.com

Abstract

Traffic simulation has a practical interest for modern society, mainly in minimizing the problems of congestion, which is the main cause of road accidents. In this way, several traffic models were developed based on Cellular Automata (CA). Moreover, the simulation of traffic system requires a high computing ability. Therefore, this paper explores the ways to use the graphics-processing unit (GPU) for reducing time simulation of such applications. We propose several software implementations to maximize GPU performance for NaSch model parallelization. Compared to CPU, the simulation results show that we need optimization effort to get better performance results for GPU.