Kitware is working with Los Alamos National Labs on a project that will extend ParaView to enable interactive visualization, even when the data size exceeds the available memory resources. Massive datasets can make visualization difficult or impossible, even with access to large parallel processing compute resources. One solution is streaming: to divide a dataset into smaller pieces, and run each of these pieces through a given visualization pipeline in turn to generate a complete result while maintaining a small memory foot print. In the paper “A modular extensible visualization system architecture for culled prioritized data streaming” published in The Proceedings of SPIE Volume 6495, Ahrens et. al describe extensions to VTK that improve upon its support for streaming. The new techniques include discarding pieces of the input dataset that do not contribute to the final visualization and prioritizing those that do. With prioritization, the most important pieces are rendered first, so that results are obtained in a significantly shorter time than with standard streaming. The new work is integrated into a custom visualization application built upon the ParaView infrastructure. This application, called StreamingParaView will be available in the upcoming ParaView 3.6 release.