Dereceli gelişen ışıma için paralel işleme

ABSTRACT PARALLEL PROCESSING FOR PROGRESSIVE REFINEMENT RADIOSITY Tolga K. Çapın M.S. in Computer Engineering and Information Science Advisor: Asst. Prof. Cevdet Aykanat September, 1993 Progressive refinement radiosity is an increasingly popular method for re alistic image synthesis of non-existing environments. The method successfully approximates the light distribution in an environment, however it requires excessive amount of computation. In this thesis, the progressive refinement method is investigated for parallelization on ring and hypercube-connected multicomputers. Two different approaches for parallelization, based on syn chronous parallelism with static task assignment, are proposed, in order to achieve better coherence in parallel light distributions and obtain good perfor mance on simple topologies. Efficient global circulation schemes are proposed in order to decrease the total volume of communication by asymptotical fac tors. The first scheme for parallelization is a modification of the sequential algorithm in that several patches shoot their energy at a time, while the sec ond scheme is based on the parallelism level of one shooting patch at a time. The proposed parallel algorithms are evaluated theoretically and implemented for ring and hypercube-connected topologies on Intel's iPSC/2 multicomputer. Load balance quality of the proposed schemes are evaluated experimentally. Keywords: Realistic Image Synthesis, Parallel Computing, Multicomputers, Radiosity, Progressive Refinement Radiosity, Ring Interconnection Topology, Hypercube Interconnection Topology. m