Mesoscopic traffic simulation on CPU/GPU

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 highperformance 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.

Masivo: Parallel Simulation Model Based on OpenCL for Massive Public Transportation Systems’ Routes

“…This implementation was tested in a road network of 10,201 nodes, 20,100 unidirectional links, and 100,000 vehicles during 1000 simulation ticks (each tick was 1 s). The simulation time for this network was 4720 ms for CPU and 423 ms for GPU, getting a speed-up of 11.2 and a real-time factor of 2364 [60].…”

“…In 2017, Song et al implemented the mesoscopic traffic simulation on GPU developed in [60] for a real-world scenario. The scenario was the Singapore expressway system, which is made up of 3179 nodes, 3388 links, and 9419 lanes with a demand modeled as trips from 4106 OD pairs.…”

“…For these reasons, novel works have developed new models and simulators that employ parallel architectures to get the simulation results more efficiently. Currently, we find mesoscopic [13,[56][57][58][59][60] and microscopic [61] traffic simulators based in GPUs (Graphics Processing Units), HPC (High Performance Computing) [62][63][64], and microscopic simulation based on Apache Spark [55,65]. Nevertheless, these modeling and simulation approaches do not consider each passenger as a unique entity in the simulation model.…”

An alternative approach for collaborative simulation execution on a CPU+GPU hybrid system

“…In contrast, our work distributes LPs to their suitable processors in advance, so the events have their own destinations and the data to store states does not need to move. In Xu et al, 13 the authors proposed a timestepped simulation framework to run a mesoscopic traffic application on a CPU + GPU hybrid system. The framework was demonstrated to simulate 100,000 vehicles moving on a large-scale grid road network.…”

Load-balancing and synchronisation for parallel agent-based road traffic simulation

“…It is able to make short-term traffic predictions in real-time, and thus is appropriate for applications involving the evaluation of individual vehicles and operating at or faster than real-time speed [27]. To improve the execution speed of large-scale mesoscopic traffic simulations, parallel simulation techniques have also been investigated for mesoscopic traffic simulation [34,35]. However, mesoscopic models do not capture the detailed movement of vehicles (e.g., accelerate and decelerate) or the exact locations of individual vehicles in the street.…”