We have entered the Big Data era, where heterogeneous architecture and deep memory hierarchy characterize the future generation of supercomputing environment. To address the large data challenge, advanced visualization techniques and data representations are required to be adaptive to the underlying architectures and to exploit data locality for efficient memory usage across heterogeneous computing environments. These considerations motivate us to develop advanced visualization strategies that aim to enable efficient feature classification, optimal data management and high-performance computation.

In this seminar, I will first introduce novel transfer functions that assist us in exploring inherent spatial relationships of a volume data. In addition, when various large datasets are involved, we must holistically consider both the volume and variety challenges and address both the data co-location and co-alignment in a distributed environment. I will then present how we solve the efficient analysis and visualization of Big-Earth-Data represented in different data models.