Summary of the SciTech 2020 Technical Panel on In Situ/In Transit Computational Environments for Visualization and Data Analytics

摘要

At the AIAA SciTech 2020 conference, the Meshing, Visualization and Computational Environments Technical Committee hosted a special technical panel on In Situ/In Transit Computational Environments for Visualization and Data Analytics. The panel brought together leading experts from industry, software vendors, Department of Energy, Department of Defense and the Japan Aerospace Exploration Agency (JAXA). In situ and in transit methodologies enable Computational Fluid Dynamic (CFD) simulations to avoid the excessive overhead associated with data I/O at large scales especially as simulations scale to millions of processors. These methods either share the data analysis/visualization pipelines with the memory space of the solver or efficiently off load the workload to alternate processors. Using these methods, simulations can scale and have the promise of enabling the community to satisfy the Knowledge Extraction milestones as envisioned by the CFD Vision 2030 study for "on demand analysis/visualization of a 100 Billion point unsteady CFD simulation". This paper summarizes the presentations providing a discussion point of how the community can achieve the goals set forth in the CFD Vision 2030.