David M. Zareski. Parallel decomposition of view-independent global illumination algorithms. Master's thesis, Program of Computer Graphics, Cornell University, expected 1995.  Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Coarse-Grained Parallelism for Hierarchical Radiosity Using.. - Funkhouser (1996) (17 citations) (Correct)
....to the head of its own queue. Load balancing was achieved by task stealing idle processors removed and processed interactions from the tail of other processors queues. Due its communication intensive nature, this approach is not practical for a network of distributed workstations. Zareski [31] implemented a parallel version of the hierarchical radiosity algorithm on a network of workstations using a master slave architecture in which each slave performed patch ray intersection calculations for a separate subset of patches in the scene. For each element element interaction, the master ....
Zareski, D. Parallel Decomposition of View-Independent Global Illumination Algorithms. Master's thesis, Cornell University, 1996.
Efficient Parallel Global Illumination using Density.. - Zareski, Wade, Hubbard, .. (1995) (4 citations) Self-citation (Zareski) (Correct)
.... hierarchical radiosity (before the advent of clustering) using a two pass method with discontinuity meshing [17] The local pass, which refines the initial global pass solution, parallelized quite well on a small scale, with a parallel efficiency of 75 or higher for up to eight processors [34]. This was because the potentially O(n 2 ) interactions between patches during the local pass was significantly reduced using visibility determination preprocessing techniques [30] We found no clear way to parallelize the global pass, however. Since the global pass takes about one quarter of ....
....Figure 8: Parallel density estimation and meshing speedup for the furniture warehouse. Execution time ranged from a maximum of 9h59m to a minimum of 2h03m. did not noticeably affect the parallel performance. Additional results and further analysis of this parallel implementation are also available [34]. 7 Conclusion and Future Work We have presented a parallel version of the recently introduced Density Estimation global illumination algorithm [23] that has a straightforward implementation and provides reasonably good parallel performance, at least on a small scale. The most obvious bottleneck ....
David M. Zareski. Parallel decomposition of view-independent global illumination algorithms. Master's thesis, Program of Computer Graphics, Cornell University, expected 1995.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC