| Contents |
| Coarse-graining is ubiquitous in many application fields, including molecular dynamics, complex fluids, materials science, ... Starting from a fine model, one derives a coarse model, which is a good approximation of the fine model in some regimes. The coarse model is cheaper to simulate than the fine one, because it typically implies a simpler physics, and/or the dimensionality of the state variable is smaller.
We will review recent advances using the parareal algorithm to address this context. The algorithm uses the coarse model as a predictor (which is iteratively corrected) to efficiently compute the trajectory of the fine model, using parallel-in-time computations. |
|