Cache-oblivious mesh layouts

Abstract

We present a novel method for computing cache-oblivious layouts of large meshes that improve the performance of interactive visualization and geometric processing algorithms. Given that the mesh is accessed in a reasonably coherent manner, we assume no particular data access patterns or cache parameters of the memory hierarchy involved in the computation. Furthermore, our formulation extends directly to computing layouts of multi-resolution and bounding volume hierarchies of large meshes. We develop a simple and practical cache-oblivious metric for estimating cache misses. Computing a coherent mesh layout is reduced to a combinatorial optimization problem. We designed and implemented an out-of-core multilevel minimization algorithm and tested its performance on unstructured meshes composed of tens to hundreds of millions of triangles. Our layouts can significantly reduce the number of cache misses. We have observed 2−20 times speedups in view-dependent rendering, collision detection, and isocontour extraction without any modification of the algorithms or runtime applications.

RT-DEFORM: Interactive Ray Tracing of Dynamic Scenes using BVHs

“…We would like to extend our algorithm to handle larger deforming models, which would require more efficient or localized update methods that only change the parts of the hierarchy that have deformed since the last frame. Also, we would like to investigate cache-coherent layout computation methods [42][43][44] of deforming models in order to efficiently handle them. Another interesting problem is the better use of multiprocessor architectures in the context of hierarchy construction and updates.…”

Single-strips for fast interactive rendering

“…Extending this concept, the efficient use of extra vertex registers has not only been exploited in geometry compression [9,6] for bandwidth reduction, but also for improved rendering [21,5,38]. In [21], multi-register vertex caching is used to increase the locality of vertex references, which reduces geometry transfer and transform costs significantly.…”

“…In [21], multi-register vertex caching is used to increase the locality of vertex references, which reduces geometry transfer and transform costs significantly. Similarly to our approach, [38] construct triangle strips that enhance the use of the vertex cache, regardless of its size. Further, [5,36] also explore the relationship between mesh locality and triangle strips.…”

Comparative Analysis of Triangulation Libraries for Modeling Large Point Clouds from Land and Their Infrastructures

“…More specifically, this strategy splits the original point cloud into small datasets, then recursively merges their triangulations to complete the Delaunay triangulation. The merge of subsets and the modification of overlap areas is the most crucial step in the process [16]; (ii) Cache-efficient strategies, which act directly in the hardware memory (caches and virtual memory) [17]. These strategies can respond to optimized software for a particular cache architecture or software designed to cooperate well with any cache or virtual memory, regardless of the details of its architecture [18]; (iii) External memory strategies, which outsource the calculation through data structures stored on a disk [19].…”