Particle tracing for flow visualization is used to generate streamlines and pathlines. It is difficult to parallelize efficiently because of widely varying communication and computation loads. Examples are shown below.
The data structures are designed for exchange of particles in 4D time-space neighborhoods, as shown below. Further reading: Peterka, T., Kendall, W., Goodell, D., Nouanesengsey, B., Shen, H.-W., Huang, J., Moreland, K., Thakur, R., Ross, R.: Performance of Communication Patterns for Extreme-Scale Analysis and Visualization. Journal of Physics: Conference Series SciDAC 2010, 2010. pdf bibtex
Three approaches to load balancing are shown below: particle termination, multiblock assignment, and dynamic block re-assignment. Further reading: Peterka, T., Ross, R., Nouanesengsey, B., Lee, T.-Y., Shen, H.-W., Kendall, W., Huang, J.: A Study of Parallel Particle Tracing for Steady-State and Time-Varying Flow Fields. Proceedings IPDPS'11, Anchorage AK, May 2011. pdf bibtex
