The Challenges of In Situ Analysis for Multiple Simulations

Abstract

In situ analysis and visualization have mainly been applied to the output of a single large-scale simulation. However, topics involving the execution of multiple simulations in supercomputers have only received minimal attention so far. Some important examples are uncertainty quantification, data assimilation, and complex optimization. In this position article, beyond highlighting the strengths and limitations of the tools that we have developed over the past few years, we share lessons learned from using them on large-scale platforms and from interacting with end users. We then discuss the forthcoming challenges, which future in situ analysis and visualization frameworks will face when dealing with the exascale execution of multiple simulations.