A survey of parallel particle tracing algorithms in flow visualization

Abstract

Particle tracing is a very important method in flow field data visualization and analysis. By placing particle seeds in the flow domain and tracing the trajectory of each particle, users can explore and analyze the hidden local or global features in the flow field. However, particle tracing is computational complex and intensive. As the size and complexity of data continue to increase, tracing particles efficiently through parallel computing for flow field visualization and analysis becomes a popular trend in recent years. In this paper, we summarize the state-of-the-art researches on parallel particle tracing algorithms in flow visualization. According to the problems and challenges in the parallelization of particle tracing, methods are divided into three categories, including task parallelism, data parallelism, and hybrid methods that combine task and data parallelism. We show the pros and cons of these algorithms and their relationships for summarization. At the end of this survey, we also look into the research trends and discuss the remaining challenges for the possible future work.