Abstract
Based on cluster, a modified parallel computation model MSG (Master-Slave-Gleaner) model is presented. MSG model can be used as a parallel computation model to develop high performance volume rendering algorithms. A very complex parallel splatting algorithm was used to test the efficiency and the practicability of the MSG model in IBM Cluster 1350. The results of theoretical analysis and numerical testing showed that the modified computing model presented can decrease computing time, increase accelerator and improve computing efficiency greatly. In addition, the MSG model is suitable for a wide variety of volume rendering algorithms and has good scalability. © Springer-Verlag 2004.
Cite
CITATION STYLE
Li, X., Sun, J., Zhang, J., Qi, Z., & Li, G. (2004). A modified parallel computation model based on cluster. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3044, 252–261. https://doi.org/10.1007/978-3-540-24709-8_27
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
