Large scale transportation simulations on Beowulf Clusters

Nurhan Cetin¹  and Kai Nagel²
Dept. of Computer Science, ETH Zürich, Switzerland
Postal: ETH Zentrum IFW B27.1, CH-8092 Zürich, Switzerland

Abstract:

If the individual entities in a system are used as the main components of a traffic simulation, the simulation is called microscopic. When the traffic density is high and the area covered is wide, the individual elements of a microscopic simulation and also the simple rules such as car following, lane changing, gap acceptance, can result in complex behaviors. Such a large scale transportation simulation can consume more time and more computing resources. A parallel computing approach to such a big traffic system might be economical and efficient in terms of money and consumed resources. This paper describes a parallel approach to a microscopic traffic simulation. The parallelization method is domain decomposition, which means that each CPU of the parallel computer is responsible for a different geographical area of the simulated region. We describe how information between domains is exchanged, and how the transportation network graph is partitioned. An adaptive scheme is used to optimize load balancing. We demonstrate how computing speeds of a parallel micro-simulations can be systematically predicted once the scenario and the computer architecture are known. This makes it possible, for example, to decide if a certain study is feasible with a certain computing budget, and how to invest that budget. The main ingredients of the prediction are knowledge about the parallel implementation of the micro-simulation, knowledge about the characteristics of the partitioning of the transportation network graph, and knowledge about the interaction of these quantities with the computer system. In particular, we investigate the differences between switched and non-switched topologies, and the effects of 10 Mbit, 100 Mbit, and Gbit Ethernet.

Keywords: Parallel computing, traffic simulation, transportation planning