Accelerated isosurface extraction in time-varying fields

Abstract

ÐFor large time-varying data sets, memory and disk limitations can lower the performace of visualization applications. Algorithms and data structures must be explicitly designed to handle these data sets in order to achieve more interactive rates. The Temporal Branch-on-Need Octree (T-BON) extends the three-dimensional branch-on-need octree for time-varying isosurface extraction. This data structure minimizes the impact of the I/O bottleneck by reading from disk only those portions of the search structure and data necessary to construct the current isosurface. By performing a minimum of I/O and exploiting the hierarchical memory found in modern CPUs, the T-BON algorithm achieves high performance isosurface extraction in time-varying fields. This paper extends earlier work on the T-BON data structure by including techniques for better memory utilization, out-of-core isosurface extraction, and support for nonrectilinear grids. Results from testing the T-BON algorithm on large data sets show that its performance is similar to that of the three-dimensional branch-on-need octree for static data sets while providing substantial advantages for timevarying fields. Index TermsÐIsosurface, time-dependent scalar field visualization, multiresolution methods, octree, bricking, unstructured grid visualization, out-of-core visualization. ae 1. Our test data sets range in size from 8.4MB to 537MB per time step.

Information-Aware 2^n-Tree for Efficient Out-of-Core Indexing of Very Large Multidimensional Volumetric Data

“…However, it can not efficiently support the slicing process since cells are organized by their interval values. Sutton and Hansen [3] introduce the Temporal Branch-on-Need-octree (T-BON) to extract isosurfaces for each time step separately. Another related work is the PHOT data structure developed in [11].…”

“…A typical way of building indexing structures in the case of time-varying volumetric data such as the one described above is to build a separate indexing structure on each time step of the data set. For example, Sutton and Hansen's temporal branch-on-need structure (T-BON) [3] is the most representative. Their strategy is to build an out-of-core version of Branch-On-Need-Octree (BONO) [4], in which each leaf node is of disk page size, for each time step and to store general common infrastructure of the trees in a single file.…”

A unified infrastructure for parallel out-of-core isosurface extraction and volume rendering of unstructured grids

“…Leutenegger and Ma [29] and Farias and Silva [17] developed out-of-core volume rendering approaches. Shen et al [44] and Sutton and Hansen [49] reported out-of-core visualization for time-varying datasets. For surface simplification, Hoppe [25] proposed view-dependent simplification method for terrains larger than main memory, and Lindstrom [30] gave an out-of-core technique to simplify large polygonal models, which can perform simplification very efficiently but does not produce any hierarchical structure for level-of-detail rendering.…”

“…In addition, Bajaj et al [2] proposed a parallel and out-of-core approach based on contour propagation from seed cells. Sequential isosurface techniques for time-varying data were given by Shen [43] and Sutton and Hansen [49].…”

Big Wins with Small Application-Aware Caches

“…ADR/DataCutter [3] is a middleware infrastructure based on R-trees [13] to store and retrieve large multi-dimensional spatial datasets. Similarly, various approaches in the scientific visualization community map and match dataset and work units to disk I/O blocks and use well-known indexing schemes to alleviate the I/O bottleneck [2,27,6,30]. These approaches rely on standard database buffer managers.…”