Large data volume visualization on distributed multiprocessor systems

Publisher Summary This chapter discusses the large data volume visualization on distributed multiprocessor systems. The large data visualization system described here is divided into two parts: server and client. This division allows carrying out the principle part of visualization process on supercomputer and transferring to a user's workplace only the minimum information required directly for construction of the prepared image. Such approach assumes that the image is finally formed on a user's workplace and it is possible to use modern multimedia hardware for better clarity of the visual information. In conclusion it should be noted that the suggested surface compression algorithms possess high computational efficiency especially for processing results of the computing experiments carried out on multiprocessor systems. This is because the suggested algorithms allow effective paralleling and are suitable for processing grid data where the volume is comparable with the whole operative memory of computer system.