A System Architecture for Data-Oriented Visualization

In this paper we present a system architecture which will allow several users to perform cooperative scientific visualization in a heterogeneous high-performance envi-ronment. In contrast to most available visualization systems this architecture tries to find a balanced relationship between the algorithms and the data. Comparable to the Controller which controls the execution flow storage managers are introduced which control the dataflow between modules which might reside on different computers. An overview of the system architecture will be given and an inspection of the potential of this approach will be performed.