Improving SIMD efficiency for parallel Monte Carlo light transport on the GPU
Abstract
Monte Carlo Light Transport algorithms such as Path Tracing (PT), Bi-Directional Path Tracing (BDPT) and Metropolis Light Transport (MLT) make use of random walks to sample light transport paths. When parallelizing these algorithms on the GPU the stochastic termination of random walks results in an uneven workload between samples, which reduces SIMD efficiency. In this paper we propose to combine stream compaction and sample regeneration to keep SIMD efficiency high during random walk construction, in spite of stochastic termination. Furthermore, for BDPT and MLT, we propose to evaluate all bidirectional connections of a sample in parallel in order to balance the workload between GPU threads and improve SIMD efficiency during sample evaluation. We present efficient parallel GPU-only implementations for PT, BDPT, and MLT in CUDA. We show that our GPU implementations outperform similar CPU implementations by an order of magnitude.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
GPU parallelism boosts the speed of the algorithm's SIMD-based components relative to older methods.
“The involvement of GPU in the algorithm allows it to make use of the parallel-computing architecture. This improves the performance of the paradigm’s parts that follow the SIMD structure, with respect to the traditional approaches that have been employed to solve the same problem. (Antwerpen 2011; Zafar et al. 2010)”
From GPU-Accelerated implementation of a genetically optimized image encryption algorithm (2021) · cites this paper as Antwerpen 2011
Cited by
Sorted by citations
GPU-Accelerated implementation of a genetically optimized image encryption algorithm
Bharadwaj, B., Saira Banu, J., Madiajagan, M., Ghalib, M. R., Castillo, O., & Shankar, A. (2021) · Soft Computing · doi:10.1007/s00500-021-06225-y · Cited by 14 · see citing works
“(Antwerpen 2011; Zafar et al. 2010) The paper also presents a genetic optimization exercise in detail, that was performed to enhance the security metrics of the PRNG, by the process of hyper-parameter tuning.”
BackgroundMethodEstimation of Volume Rendering Efficiency with GPU in a Parallel Distributed Environment
Monte, C. F. P., Piccoli, F., Luciano, C., Rizzi, S., Bianchini, G., & Scutari, P. C. (2013) · Procedia Computer Science · doi:10.1016/j.procs.2013.05.307 · Cited by 11 · see citing works
- Authors
- van Antwerpen, D.
- Affiliations
- Delft University of Technology, Netherlands
- Venue
- Proceedings of the ACM SIGGRAPH Symposium on High Performance Graphics
- Publisher
- ACM
- Published
- 2011-08-05
- Pages
- 41-50
- OpenAlex
- W2009173369
- Cited by
- 40
- References
- 0
- Citations by year
- 2025: 1 · 2024: 2 · 2023: 1 · 2021: 3 · 2020: 2 · 2018: 3 · 2017: 4 · 2016: 2 · 2015: 3 · 2014: 4 · 2013: 9 · 2012: 5
- FWCI
- 6.98
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- High Performance Graphics (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Graphics and Computer-Aided Design › Computer Graphics and Visualization Techniques
- Physical Sciences › Computer Science › Hardware and Architecture › Parallel Computing and Optimization Techniques
- Physical Sciences › Engineering › Computational Mechanics › Radiative Heat Transfer Studies
- Keywords
- Monte Carlo light transport, GPU rendering, path tracing, bidirectional path tracing, Metropolis light transport, load balancing, CUDA
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
