Large Scale Satellite Imagery Simulations with Physically Based Ray Tracing on Tianhe-1A Supercomputer

Wu, Chih Yang; Zhang, Yunquan; Yang, Congli

doi:10.1109/hpcc.and.euc.2013.84

Cited by 4 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unfortunately, currently available demand-driven assignment partitioning algorithm such as those in [10], [17], [18] doesn't work for the MLT algorithm, largely because a demand-driven method subdivides the image plane into a number offixed sub-regions for compute nodes, thus rays going through pixels of a given fixed sub region are tested for intersections with the objects, i.e., we need to know the locations of pixels of the given assignments. However, the MLT algorithm stochastically generates a sample path from the initial sample, i.e., the pixel location is produced randomly, which indicates a demand-driven method doesn't work for the MLT algorithm.…”

Section: Parallelization St Rategiesmentioning

confidence: 99%

“…In a previous work [9], we have developed a physically based ray tracing algorithm running on a small cluster with many hundred CPU cores. Although the work doesn't involve the MLT algorithm, it initiates our another work [10] for a scalable physically based ray tracing algorithm for satellite imagery simulations using up to 16,800 CPU cores on the Tianhe-IA supercomputer.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Physically based parallel ray tracer for the Metropolis light transport algorithm on the Tianhe-2 supercomputer

Zhang

Yang

et al. 2014

2014 20th IEEE International Conference on Parallel and Distributed Systems (ICPADS)

Self Cite

View full text Add to dashboard Cite

Developing an efficient and highly scalable ray tracer for the Metropolis light transport algorithm is becoming increasingly important as the request for photorealistic images becomes a common trend. Although the Metropolis light trans port algorithm has produced some of the most realistic images to date, it usually takes a great amount of time to render an image. The development of an efficient and highly scalable ray tracer for the Metropolis light transport algorithm is hard due in large part to the irregular memory access patterns, the imbalanced workload of light-carrying paths and the complicated mathematical model and complex physical processes.In this paper, we present a highly scalable physically based parallel ray tracer for the Metropolis light transport algorithm.Firstly, we present the idea of snapshot and sub-snapshot, then propose a novel assignment partitioning algorithm for compute nodes and CPU cores since the demand-driven assignment partitioning algorithms don't work. Secondly, we propose a physically based parallel ray racing framework for the Metropolis light transport algorithm, which is based on a master-worker architecture. Finally, we discuss the issue of granularity of the assignment partitioning and some optimization strategies for improving overall performance, then a hybrid scheduling strategy combining a static and dynamic scheduling strategy is described.Experiments show that our physically based ray tracer almost reaches linear speedup by using 26,400 CPU cores on the Tianhe-2 supercomputer. Our ray tracer is more efficient and highly scalable.Keywords-physically based ray tracing, bidirectional path trac ing, Metropolis light transport algorithm, assignment partition ing, hybrid scheduling, distributed computing.

show abstract

Section: Parallelization St Rategiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%