Yan Xu
ABSTRACT
Mesoscopic traffic simulation is an important branch of technology to support offline large-scale simulation-based traffic planning and online simulation-based traffic management. One of the major concerns using mesoscopic traffic simulations is the performance, which means the required time to simulate a traffic scenario. At the same time, the GPU has recently been a success, because of its massive performance compared to the CPU. Thus, a critical question is "whether the GPU can be a potential high-performance platform for mesoscopic traffic simulations"? To the best of our knowledge, there is no clear answer in the research area. In this paper, we firstly propose a comprehensive framework to run a traditional time-stepped mesoscopic traffic simulation on CPU/GPU. Then, we design a boundary processing method to guarantee the correctness of running mesoscopic supply traffic simulations on the GPU. Thirdly, the proposed mesoscopic traffic simulation framework is demonstrated to simulate 100,000 vehicles moving on a large-scale grid road network. In this case study, running a mesoscopic supply traffic simulation on a GPU (GeForce GT 650M) gives 11.2 times speedup, compared with running the same supply simulation on a CPU core (Intel E5-2620). In the end, this paper explains the theoretical limitation of running mesoscopic supply traffic simulations on the GPU. In conclusion, regardless of high system complexity, the proposed mesoscopic traffic simulation framework on CPU/GPU provides an innovative and promising solution for high-performance mesoscopic traffic simulations.
- Barcelo J. (editor), "Fundamentals of Traffic Simulation", International Series in Operations Research & Management Science, 2010, Springer, New York.Google Scholar
- Yang W., "Scalability of Dynamic Traffic Assignment", Ph.D. thesis, Massachusetts Institute of Technology, 2009.Google Scholar
- Ben-Akiva, M., Gao, S., Wei, Z. and Yang, W., "A dynamic traffic assignment model for highly congested urban networks", Transportation Research Part C, 2012, 24: 62--82.Google ScholarCross Ref
- Ben-Akiva, M., Bierlaire, M., Burton, D., Koutsopoulos, H. N., and Mishalani, R., "Network state estimation and prediction for real-time traffic management", Networks and Spatial Economics, 2001, 1(3/4):293--318.Google ScholarCross Ref
- Mahmassani H.S., Hu T., and Jaykrishnan R., "Dynamic traffic assignment and simulation for advanced network informatics (DYNASMART)", Proceedings of the 2nd International Capri Seminar on Urban Traffic Networks, Capri, Italy, 1992.Google Scholar
- Ziliaskopoulos A. K., Waller S. T., Li Y., and Byram, M., "Large-scale dynamic traffic assignment: Implementation issues and computational analysis", Journal of Transportation Engineering, 2004, 130(5): 585--593.Google ScholarCross Ref
- Gordon D. B. C. and Gordon I. D. D., "Paramics - Parallel Microscopic Simulation of Road Traffic". The Journal of Supercomputing, 1996, pp. 25--53.Google Scholar
- Çetin, N., "Large-scale parallel graph-based simulations", Ph.D. thesis, ETH Zurich, Switzerland, 2005.Google Scholar
- Aydt H., Yadong X., Michael L., and Alois K., "A Multi-threaded Execution Model for the Agent-Based SEMSim Traffic Simulation", Proceedings of AsiaSim 2013.Google ScholarCross Ref
- Nvidia, "CUDA C Programming Guide", 2013, available at http://www.nvidia.com/CUDA.Google Scholar
- Michalakes J. and Vachharajani M., "GPU acceleration of numerical weather prediction", IPDPS 2008: IEEE Int'l Symp. Parallel and Distributed Processing, 2008.Google ScholarCross Ref
- Buck I., "GPU computing with NVIDIA CUDA". In SIGGRAPH '07, New York, NY, USA, 2007. Google ScholarDigital Library
- Fan, Z., Qiu, F., Kaufman, A., and Yoakum-Stover, S., "GPU cluster for high performance computing". In Proceedings of SC'04, 2004. Google ScholarDigital Library
- Perumalla, K. S., Brandon G. A., Srikanth B. Y., and Sudip K. S., "GPU-based real-time execution of vehicular mobility models in large-scale road network scenarios", International Workshop on Principles of Advanced and Distributed Simulation, 2009. Google ScholarDigital Library
- Strippgen, D. and Nagel, K., "Multi-agent traffic simulation with CUDA", Proceedings of International Conference on High Performance Computing & Simulation, Leipzig, 2009.Google ScholarCross Ref
- Yan X., Xiao S., Zhiyong W. and Gary T., "An Entry Time based Supply Framework (ETSF) for Mesoscopic Traffic Simulations", submitted to Simulation Modelling Practice and Theory, 2014.Google Scholar
- Denis G., Jose-Juan T., Samuel A. and Roshan M. D. S., "Graphics processing unit based direct simulation Monte Carlo", Simulation: Transactions of the Society for Modeling and Simulation International, 2012, 88(6): 680--693. Google ScholarDigital Library
- Brodtkorb A. R., Trond R. H. and Martin L. S., "Graphics processing unit (GPU) programming strategies and trends in GPU computing", Journal of Parallel and Distributed Computing, 2013, 73: 4--13. Google ScholarDigital Library
- Perumalla, K. S., "Discrete Event Execution Alternatives on Gen-eral Purpose Graphical Processing Units (GPGPUs)", International Workshop on Principles of Advanced and Distributed Simulation, 2006. Google ScholarDigital Library
- Passerat-Palmbach, J.; Mazel, C.; Hill, D. R C, "Pseudo-Random Number Generation on GP-GPU," International Workshop on Principles of Advanced and Distributed Simulation, 2011. Google ScholarDigital Library
- Xiaosong L., Wentong C. and Stephen J. T., "GPU Accelerated Three-stage Execution Model for Event-parallel Simulation", International Workshop on Principles of Advanced and Distributed Simulation, 2013. Google ScholarDigital Library
- Park H. and Fishwick P. A., "An analysis of queuing network simulation using GPU-based hardware acceleration", ACM Transactions on Modeling and Computer Simulation. 2011, 21(3). Google ScholarDigital Library
Index Terms
- Mesoscopic traffic simulation on CPU/GPU
Recommendations
Radiation modeling using the Uintah heterogeneous CPU/GPU runtime system
XSEDE '12: Proceedings of the 1st Conference of the Extreme Science and Engineering Discovery Environment: Bridging from the eXtreme to the campus and beyondThe Uintah Computational Framework was developed to provide an environment for solving fluid-structure interaction problems on structured adaptive grids on large-scale, long-running, data-intensive problems. Uintah uses a combination of fluid-flow ...
On the Efficacy of a Fused CPU+GPU Processor (or APU) for Parallel Computing
SAAHPC '11: Proceedings of the 2011 Symposium on Application Accelerators in High-Performance ComputingThe graphics processing unit (GPU) has made significant strides as an accelerator in parallel computing. However, because the GPU has resided out on PCIe as a discrete device, the performance of GPU applications can be bottlenecked by data transfers ...
Adaptive Optimization for Petascale Heterogeneous CPU/GPU Computing
CLUSTER '10: Proceedings of the 2010 IEEE International Conference on Cluster ComputingIn this paper, we describe our experiment developing an implementation of the Linpack benchmark for TianHe-1, a petascale CPU/GPU supercomputer system, the largest GPU-accelerated system ever attempted before. An adaptive optimization framework is ...
Comments