109/04/16 Large-scale Scientific Data Processing and Visualization
Abstract: As the advance of large-scale supercomputers enable scientists to model complex physical phenomena and produce a huge amount of datasets, handling these datasets to understand the science becomes a daunting challenge.
In this talk, I will discuss how visualization helps scientific data understanding and the scheme to handle large-scale simulation datasets. The first part of the talk will briefly introduce the connections between visualization and data analysis. In the second part of the talk,
In this talk, I will discuss how visualization helps scientific data understanding and the scheme to handle large-scale simulation datasets. The first part of the talk will briefly introduce the connections between visualization and data analysis. In the second part of the talk,
I will discuss the challenges of scientific data visualization and analysis in the era of big data. One challenge is the long I/O time for simulation data movement from the supercomputer to the post-analysis machine. The other one is the requirement of considerable disk space to keep all datasets for interactive data analysis. Integrating in situ data processing and machine learning becomes a promising solution for large-scale data challenges.
They have successfully demonstrated the power and flexibility to handle large datasets for data visualization and analysis from a variety of simulations. In the end, I will finish the talk with important open questions and research directions.