Immersive and Interactive Exploration of Billion-Atom Systems

Abstract

We have developed a visualization system, named Atomsviewer, to render a billion atoms from the results of a molecular dynamics simulation. This system uses a hierarchical view frustum culling algorithm based on the octree data structure to efciently remove atoms that are outside of the eld of view. A novel occlusion culling algorithm, using a probability function, then selects atoms with a high probability of being visible. These selected atoms are further tested with a traditional occlusion culling algorithm before being rendered as spheres at varying levels of detail. To achieve scalability, Atomsviewer is distributed over a cluster of PCs that execute a parallelized version of the hierarchical view frustum culling and the probabilistic occlusion culling, and a graphics workstation that renders the atoms. We have used Atomsviewer to render a billion-atom data set on a dual processor SGI Onyx2 with an In niteReality2 graphics pipeline connected to a four-node PC cluster.

A divide-and-conquer/cellular-decomposition framework for million-to-billion atom simulations of chemical reactions

“…We have developed a scalable visualization system, Atomsviewer, to allow the viewer to walk through a billion atoms [59]. The system uses: the octree data structure as an efficient abstraction mechanism to extract atoms within the field-of-view (view frustum culling); a novel probabilistic approach to remove far atoms that are hidden by other atoms (occlusion culling); parallel/distributed processing of these culling tasks on a Linux cluster connected to a graphics server; a machine-learning approach to predict the user's next movement and prefetch data from the Linux cluster to the graphics server; and multiresolution rendering.…”

Numerical Thinking in Algorithm Design and Analysis

“…We may need to deal with equations and mathematical programming that involve hundreds of millions of variables (Sharma et al 2002). We may need to analyze data and graphs such as web logs, social networks, and web graphs that are massive (e.g., of hundreds billions of nodes Gulli and Signorini 2005), complex, and dynamic.…”

Space–time multiresolution atomistic visualization of MgO and MgSiO3 liquid data

“…Current trends in such computational studies demand visualization along three major directions: First, ever-larger sets of data are produced by large-scale simulations. The challenge is simply how to graphically render such massive datasets (Sharma et al 2003). Second, a wide range of properties including structure, elasticity, rheology and melting of materials are now simulated on a routine basis.…”

“…The common examples include Molscript (Kraulis 1991), VMD (Humphrey et al 1996), XcrysDen (Kokaji 1999;2003), Atomsviewer (Sharma et al 2003), CrystalMaker (http://www.crystalmaker.com) and amiraMol (http://www.amiravis.com/mol) commercial packages, and numerous other public domain systems such as Aviz, gOpenMol, VASP DataViewer, PyMD, etc. Most systems exploit 3D graphics supported by desktop computers but some also support immersive and interactive visualization requiring specialized resources such as Immersadesk and CAVE.…”