Challenges of streaming in the visualization process

Abstract

In visualization applications, the data has been increasing its size at very big rates. Processing of the whole data in main memory becomes impossible due to size limitations. A way of dealing with this constraint is to apply streaming to the visualization. The key idea is to exploit data locality in relationships to produce constant and continuous streams of data flowing through the visualization process. Issues like data dependencies, stream re-arranging, use of progressive algorithms, etc. have to be taken in consideration. In this paper we outline the main issues derived from the application of streaming in visualization. Our main objective is their identification as a previous step to define a general streaming framework for visualization. Some of the results presented here arose during the design and development of ad-hoc prototypes that we made as an initial approximation to the problem.