The ever-increasing sizes of data produced from a variety of scientific studies post a formidable challenge for the subsequent data analysis and visualization tasks. While steady advances in graphics hardware enable faster rendering, achieving interactive visualization of large data must also rely on effective data filtering and organization. In many cases, the best interactivity can only be obtained by taking into account the intrinsic properties of the data and domain knowledge to better reduce and organize the data for visualization. As a result, in recent years, we have seen increasing research and development efforts into the area of information and knowledge assisted visualization (IKV). In this paper, we survey research in IKV of scientific data and also identify a few directions for further work in this emerging area.