{"title":"A hypercube Ray-tracer","authors":"J. Salmon, J. Goldsmith","doi":"10.1145/63047.63073","DOIUrl":null,"url":null,"abstract":"We describe a hypercube ray-tracing program for rendering computer graphics. For small models, which fit in the memory of a single processor, the ray-tracer uses a scattered decomposition of pixels to balance the load, and achieves a very high efficiency. The more interesting case of large models, which cannot be stored in a single processor, requires a decomposition of the model data as well as the pixels. We present algorithms for constructing a decomposition based upon information about the frequency with which different elements of the model are accessed. The resulting decomposition is approximately optimized to minimize communication and achieve load balance.","PeriodicalId":299435,"journal":{"name":"Conference on Hypercube Concurrent Computers and Applications","volume":"365 ","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"1989-01-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"58","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Conference on Hypercube Concurrent Computers and Applications","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/63047.63073","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 58
Abstract
We describe a hypercube ray-tracing program for rendering computer graphics. For small models, which fit in the memory of a single processor, the ray-tracer uses a scattered decomposition of pixels to balance the load, and achieves a very high efficiency. The more interesting case of large models, which cannot be stored in a single processor, requires a decomposition of the model data as well as the pixels. We present algorithms for constructing a decomposition based upon information about the frequency with which different elements of the model are accessed. The resulting decomposition is approximately optimized to minimize communication and achieve load balance.