Improved OpenCL-based Implementation of Social Field Pedestrian Model

Two aspects of improvements are proposed for the OpenCL-based implementation of the social field pedestrian model. In the aspect of algorithm, a method based on the idea of divide-and-conquer is devised in order to overcome the problem of global memory depletion when fields are of a larger size. This is of importance for the study of finer pedestrian walking behavior, which usually implies usage of large fields. In the aspect of computation, OpenCL related computation techniques are widely investigated, many of which are implemented. This includes usage of local memory, intential patch of data structures for avoidance of bank conflicts, and so on. Numerical experiments disclose that these techniques together will bring a remarkable computation performance improvement. Compared to the CPU model and the previous OpenCL-based implementation that was mainly based on the global memory, the current one can be at most 71.56 and 13.3 times faster respectively.