EG Symposium on Parallel Graphics and Visualization

Brno, Czech Republic, June 4, 2018

Keynote

Large-Scale Visualization and Multi-Resolution (GPU) Data Structures

The rapidly increasing amount of data acquired or computed in data-driven science presents a tremendous challenge to visualization and analysis. Nevertheless, for exploring, analyzing, and understanding large-scale data it is crucial to enable interactive visualization. Well-established basic approaches to tackling this challenge are on the one hand using multi-resolution representations, while often on the other hand exploiting the processing power of GPUs. However, scaling to extreme-scale data requires going beyond standard approaches. This talk will give an overview of some of our work on multi-resolution methods for very large data, such as microscopy volumes from neuroscience or large-scale particle data from molecular dynamics simulations, and designing data structures and visualization algorithms that are well-suited to the characteristics of GPU architectures.


Markus Hadwiger is an Associate Professor in computer science and the Visual Computing Center (VCC) at King Abdullah University of Science and Technology (KAUST) in Saudi Arabia, which he joined in 2009. He leads the High-Performance Visualization research group at VCC, where his research interests in the area of scientific visualization include extreme-scale visual computing and visualization, volume visualization, medical visualization, large-scale image and volume processing, multi-resolution techniques, data streaming and out-of-core processing, interactive segmentation, and GPU algorithms and architecture. He is a co-author of the book Real-Time Volume Graphics published in 2006 and has been involved in many courses and tutorials about volume rendering and visualization at ACM SIGGRAPH, ACM SIGGRAPH Asia, IEEE Visualization, and Eurographics. Prof. Hadwiger has co-authored more than 70 refereed articles.


Program

Monday, June 4

Location: Rooms 3+4

Session 1: 9:00-10:40

Opening

Direct Raytracing of Particle-based Fluid Surfaces Using Anisotropic Kernels

VisIt-OSPRay: Towards An Exascale Volume Visualization System

Robust Iterative Find-Next-Hit Ray Traversal

Coffee Break

Session 2: 11:10-12:50

Hardware-Accelerated Multi-Tile Streaming for Realtime Remote Visualization

Keynote

Lunch Break: 12:50-14:20

Session 3: 14:20-16:00

Performance-Portable Particle Advection with VTK-m

Dense Texture Flow Visualization using Data-Parallel Primitives

Revisiting the Evaluation of In Situ Lagrangian Analysis

Rapid k-d Tree Construction for Sparse Volume Data

Coffee Break

Session 4: 16:30-18:10

Interactive Visual Analysis of Multi-dimensional Metamodels

La VALSE: Scalable Log Visualization for Fault Characterization in Supercomputers

Closing