Optimizing Pairwise Box Intersection Checking on GPUs for Large-Scale Simulations

Lo, Sherman; Lee, Che-Rung; Chung, I-Hsin; Chung, Yeh‐Ching

doi:10.1145/2499913.2499918

Cited by 16 publications

(18 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…At a high level, broad‐phase algorithms can be classified by their use of spatial, temporal and sorting techniques. Spatial techniques seek to divide‐and‐conquer the problem using spatial data structures, such as Grids [LLCC13], BVHs [Cou08b], KD‐Trees [SR17], Octrees and BSP‐Trees [LCF05]. On the other hand, temporal techniques explore the behaviour of the simulation to mitigate the workload.…”

Section: Collision Detection Algorithmsmentioning

confidence: 99%

“…Multilevel Grids or unbalanced trees can also be used to more finely control the pruning. Within the GPU context, the BF approach is an important design asset due to its high predictability (which can be exploited to efficiently balance the workload across threads), and the sheer parallel power, which can drastically mitigate the quadratic nature of the algorithm [AGA12, LLCC13].…”

Section: Collision Detection Algorithmsmentioning

confidence: 99%

See 1 more Smart Citation

Broadmark: A Testing Framework for Broad‐Phase Collision Detection Algorithms

Serpa

Rodriguês

2019

Computer Graphics Forum

View full text Add to dashboard Cite

Research in the area of collision detection permeates most of the literature on simulations, interaction and agents planning, being commonly regarded as one of the main bottlenecks for large‐scale systems. To this day, despite its importance, most subareas of collision detection lack a common ground to test and validate solutions, reference implementations and widely accepted benchmark suites. In this paper, we delve into the broad‐phase of collision detection systems, providing both an open‐source framework, named Broadmark, to test, compare and validate algorithms, and an in‐deep analysis of the main techniques used so far to tackle the broad‐phase problem. The technical challenges of building this framework from the software and hardware perspectives are also described. Within our framework, several original and state‐of‐the‐art implementations of CPU and GPU algorithms are bundled, alongside three benchmark scenes to stress algorithms under several conditions. Furthermore, the system is designed to be easily extensible. We use our framework to bring out an extensive performance comparison among assembled solutions, detailing the current CPU and GPU state‐of‐the‐art on a common ground. We believe that Broadmark encompasses the principal insights and tools to derive and evaluate novel algorithms, thus greatly facilitating discussion about successful broad‐phase collision detection solutions.

show abstract

Section: Collision Detection Algorithmsmentioning

confidence: 99%

Section: Collision Detection Algorithmsmentioning

confidence: 99%

Broadmark: A Testing Framework for Broad‐Phase Collision Detection Algorithms

Serpa

Rodriguês

2019

Computer Graphics Forum

View full text Add to dashboard Cite

show abstract

“…Over the years, the continuous effort to improve CD has led to several improvements to both algorithms and spatial data structures. Many data structures have been explored, such as Grids [AGA12, LLCC13], Bounding Volumes Hierarchies (BVHs) [Cou08, LGS*09], KD‐Trees, Octrees and Binary Space Partitioning Trees (BSP‐Trees) [Gla05, LCF05, SN15, Wal07] and sorted vectors [TBW09, DSC05]. However, few works have explored the applicability of their solutions beyond simple scenarios, and often provide only basic time statistics, such as the total execution time for an evenly distributed set of randomly moving objects.…”

Section: Related Workmentioning

confidence: 99%

Flexible Use of Temporal and Spatial Reasoning for Fast and Scalable CPU Broad‐Phase Collision Detection Using KD‐Trees

Serpa

Rodriguês

2018

Computer Graphics Forum

View full text Add to dashboard Cite

Realistic computer simulations of physical elements such as rigid and deformable bodies, particles and fractures are commonplace in the modern world. In these simulations, the broad‐phase collision detection plays an important role in ensuring that simulations can scale with the number of objects. In these applications, several degrees of motion coherency, distinct spatial distributions and different types of objects exist; however, few attempts have been made at a generally applicable solution for their broad phase. In this regard, this work presents a novel broad‐phase collision detection algorithm based upon a hybrid SIMD optimized KD‐Tree and sweep‐and‐prune, aimed at general applicability. Our solution is optimized for several objects distributions, degrees of motion coherence and varying object sizes. These features are made possible by an efficient and idempotent two‐step tree optimization algorithm and by selectively enabling coherency optimizations. We have tested our solution under varying scenario setups and compared it to other solutions available in the literature and industry, up to a million simulated objects. The results show that our solution is competitive, with average performance values two to three times better than those achieved by other state‐of‐the‐art AABB‐based CPU solutions.

show abstract

“…This primitive building block has been used in a wide range of applications, such as database, statistics, artificial intelligence, image processing, simulations, etc. [3][4] [10]. With the tremendous amount of data elements to be processed, the performance of selection becomes an important factor in running these applications efficiently.…”

Section: Introductionmentioning

confidence: 99%

Efficient Algorithms for Stream Compaction on GPUs

Bakunas-Milanowski¹,

Rego

Sang

et al. 2017

IJNC

View full text Add to dashboard Cite

Stream compaction, also known as stream filtering or selection, produces a smaller output array which contains the indices of the only wanted elements from the input array for further processing. With the tremendous amount of data elements to be filtered, the performance of selection is of great concern. Recently, modern Graphics Processing Units (GPUs) have been increasingly used to accelerate the execution of massively large, data parallel applications. In this paper, we designed and implemented two new algorithms for stream compaction on GPU. The first algorithm, which can preserve the relative order of the input elements, uses a multi-level prefix-sum approach. The second algorithm, which is non-order-preserving, is based the hybrid use of the prefix-sum and the atomics approaches. We compared their performance with other parallel selection algorithms on the current generation of NVIDIA GPUs. The experimental results show that both algorithms run faster than Thrust, an open-source parallel algorithms library. Furthermore, the hybrid method performs the best among all existing selection algorithms on GPU and can be two orders of magnitude faster than the sequential selection on CPU, especially when the data size is large.

show abstract

Optimizing Pairwise Box Intersection Checking on GPUs for Large-Scale Simulations

Cited by 16 publications

References 24 publications

Broadmark: A Testing Framework for Broad‐Phase Collision Detection Algorithms

Broadmark: A Testing Framework for Broad‐Phase Collision Detection Algorithms

Flexible Use of Temporal and Spatial Reasoning for Fast and Scalable CPU Broad‐Phase Collision Detection Using KD‐Trees

Efficient Algorithms for Stream Compaction on GPUs

Contact Info

Product

Resources

About