Improving Smart City Simulation Performance with SimEDaPE and Parallelism

• Francisco Wallison Rocha USP
• Emilio Francesquini UFABC
• Daniel Cordeiro USP

DOI: https://doi.org/10.5753/wperformance.2022.223235

Resumo

In the context of smart cities, the use of simulators such as the InterSCSimulator to study urban mobility problems is increasingly frequent. However, InterSCimulator has limitations in its performance to test large scenarios such as the city of São Paulo. SimEDaPE was proposed to improve the performance of this simulator, but there are still some performance bottlenecks, such as the mapping step, that could be tackled. This work proposes two parallel implementations to improve the performance of this step. Experimental results show that the best parallel approach using 8 cores results in a 3× speedup when compared to the best sequential implementation.

Palavras-chave: Smart City, Simulations, SimEDaPE, Parallelism

Referências

Berndt, D. J. and Clifford, J. (1994). Using dynamic time warping to find patterns in time series. In Proceedings of the 3rd International Conference on Knowledge Discovery and Data Mining, AAAIWS’94, page 359–370, Seattle, WA. AAAI Press.

Dalcin, L., Bradshaw, R., Smith, K., Citro, C., Behnel, S., and Seljebotn, D. (2011). Cython: The best of both worlds. Computing in Science & Engineering, 13(02):31–39.

Hamerly, G., Perelman, E., Lau, J., and Calder, B. (2005). Simpoint 3.0: Faster and more flexible program phase analysis. Journal of Instruction Level Parallelism, 7(4):1–28.

Martins, T. G., Lago, N., de Souza, H. A., Santana, E. F. Z., Telea, A., and Kon, F. (2020). Visualizing the structure of urban mobility with bundling: A case study of the city of São Paulo. In Anais do IV Workshop de Computação Urbana, pages 178–191. SBC.

Mattson, T. G., Anderson, T. A., and Georgakoudis, G. (2021). Pyomp: Multithreaded parallel programming in python. Computing in Science Engineering, 23(6):77–80.

Paparrizos, J. and Gravano, L. (2015). K-shape: Efficient and accurate clustering of time series. In Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data, SIGMOD ’15, page 1855–1870, New York, NY, USA. Association for Computing Machinery.

Rocha, F. W., Francesquini, E., and Cordeiro, D. (2022). Fast SimEDaPE: Simulation estimation by data patterns exploration. 13ª Escola Regional de Alto Desempenho de São Paulo. To publish.

Rocha, F. W., Fukuda, J. C., Francesquini, E., and Cordeiro, D. (2021). Accelerating smart city simulations. Latin America High Performance Computing Conference. To publish.

Santana, E. F. Z., Lago, N., Kon, F., and Milojicic, D. S. (2017). InterSCSimulator: Large-scale traffic simulation in smart cities using erlang. In International Workshop on Multi-Agent Systems and Agent-Based Simulation, pages 211–227. Springer.