Parallel volume visualization on a hypercube architecture
C. Montani, Raffaele Perego, Roberto Scopigno · 1992
A parallel solution to the visualiaation of high resolution uol-ume data is presented. Baaed on the ray tracing (RT) uiau-alization technique, the system works on a distributed mem-ory MIMD architecture. A hybrid strategy to my tracing parallelitation is applied, using ray-dataflow within an im-age partition approach. This strategy allows the flexible and efiectiue management of huge dataset on architectures with limited local memory. The dataaet is distributed over the nodes using a slice-partitioning technique. The simple data partition chosen implies a atraighforward communications pattern of the visualization processes and this improves both software design and eficiency, while providing deadlock pre-vention. The partitioning technique used and the network in-terconnection topology allow for the efjicient implementation of a statical load balancing technique through pre-rendering of a low resolution image. Details related to the practical issues involved in the parallelitation of volumetric RT are discussed, with particular reference to deadlock and termi-nation issues.